Determinants of homeownership in VietNam Luận văn thạc sĩ

It is found that the household characteristics such as income, the number of members and assets of the household are statistically significant effect on the probability of homeownership.

VIETNAM- NETHERLANDS PROGRAMME FORMA IN DEVELOPMENT ECONOMICS

Determinants of Homeownership in

Vietnam

A thesis submitted in partial fulfilment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

By NGUYEN NGOC TUAN

BO GIAO DUC £lAO "

VIETNAM - NETHERLANDS PROGRAMME FORMA IN DEVELOPMENT ECONOMICS

Determinants of Homeownership in

Vietnam

A thesis submitted in partial fulfilment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

Academic Supervisor: DR HA THUC VIEN Student: NGUYEN NGOC TUAN

HO CHI MINH CITY, DECEMBER 2009

ACKNOWLEDGEMENTS

First of all, I would like to thank Dr Ha Thuc Vien, my supervisor, for his valuable comments and assistance Without his support and encouragement, my thesis would not be finished

I also thank to my classmate Mr Nguyen Thanh Danh who advised me c:in data processing for the thesis

I would like to special thank Assoc Prof Dr Nguyen Trong Hoai who has often urged and encouraged me to complete my thesis

Supports from Vietnam-Netherlands Programme for MA m Development Economics' staff and classmates are fully acknowledged

My heartfelt gratitude also goes to my family, particularly my wife and children who are my love and motivation during the studying time

HCMC, Winter 2009

NGUYEN NGOC TUAN

DECLARATION

I declare that "Determinants of Homeownership in Vietnam" is my own work,

that it has not been submitted to any degree or examination at any other universities, and that all the sources used or quoted are indicated and acknowledged by complete references

HCMC, December 2009

NGUYEN NGOC TUAN

ABSTRACT

Homeownership is considered as one of most essential needs for people living in the country It reflects the quality of life through the dwelling condition The percentage

of households that own private houses is increasing year to year Furthennore, Vietnam is

a developing country that is targeting to a market-oriented economy This has created many opportunities for improving its whole economy and society Consequently, the probability ofhomeownership is improved, as well However, the rate of poor households

is still rather high in Vietnam This may leads to many difficulties in obtaining satisfaction from homeownership Therefore, this study is to examine determinants of homeownership of households in Vietnam on the basic of descriptive statistics and econometrics It is found that the household characteristics such as income, the number of members and assets of the household are statistically significant effect on the probability

of homeownership Also, age and gender of the household heads are found to be significantly correlated its homeownership However, marital status and education attainment of the head are insignificant Other dete1minants are social-economic characteristics such as credit assistant program, immigration and employment The result

is that, immigration and employment have significant effects on homeownership probability, excluding credit assistant programs

TABLE OF CONTENTS

PAGE

ACKNOWLEDGEMENT i

DECLARATION ii

ABSTRACT iii

TABLE OF CONTENTS iv

LIST 0 F TABLES vi

LIST OF FIGURES vi

LIST OF BOXES • vi

ABBREVIATIONS AND GLOSSARIES vii

CHAPTER 1: INTRODUCTION 1

1.1 Problem statement 1

1.2 Objective of study 2

1.3 Research questions 3

1.4 Research methodology and scope of research 3

1.5 Structure ofthesis 4

CHAPTER II: LITERATURE REVIEW 5

2.1 Theoretical literature 5

2.1.1 Basic terms and concepts 5

2.1.2 Assessment of economic theoretical approaches 7

2.2 Empirical studies 16

CHAPTER III: RESEARCH METHODOLOGY 21

3.1 Analytical framework 21

3.1.1 Qualitative choice models for durable goods 21

3 1.2 Methods of homeownership research 22

3.1.3 Major determinants ofhomeownership 24

3.2 Research method 29

3.2.1 Justification of models 29

3.2.2 Econometric model specification 30

CHAPTER IV: AN OVERVIEW OF HOMEOWNERSHIP AND HOUSING SECTOR IN VIETNAM 35

4.1 An overview ofhomeownership in Vietnam 35

4.2 Changing in housing policies 39

4.3 Challenges ofhomeownership in the new economy 41

4.4 Housing demand in Vietnam: An example from Hanoi and Ho Chi Minh City .43

CHAPTER V: ECONOMETRIC ANALYSIS AND RESULTS 47

5.1 Data set 47

5.2 Logistic model ofhomeownership 49

5.3 Logistic regression results 50

5.4 Measurement of goodness-of-fit 52

5.5 Result interpretation 53

5.6 Summary 56

CHAPTER VI: CONCLUSIONS AND POLICY IMPLICATIONS S8 6.1 Conclusions 58

6.2 Policy implications , 59

REFERENCES 60

APPENDIX 65

LIST OF TABLE

Table 3.1: Variable descriptions 34 Table 4.1: Mean of determinants ofhomeownership 36 Table 4.2: Group of determinants by binary characteristics 38

ABBREVIATIONS AND GLOSSARIES

: Public Use Micro-data Sample : Statistics/Data Analysis

:United States : United Nations Development Program : Vietnam Household Living Standard Survey : Wage setting

:World-Trade-Organization

In 1991 Vietnamese people are allowed to have the right of owning a house by an ordinance on housing Therefore, homeownership is a wish of Vietnamese people (Geertman 2003) In according with economic growth, homeownership has become a proper demand in the life for a household because it is related to financial benefits (possible capital gains by selling a house later) and non-monetary benefits (physical health and development of children's cognitive capacities) in the future (Leppel 2007) In the perception of home owners, homeownership is embodied the personal success and happiness (Dietz & Haurin 2003) In addition, homeownership also increases local community stability and commitment in making a choice to live in where a household owns a house (Ratledge 1999) As a result, homeownership increases the demography stability in comparison with house renters that often change their accommodations (Dietz

& Haurin 2003) In "Urban Development and Housing Sector in Viet Nam" research, Mizuno and Ishigami (1999) found that households do not like to rent a house for their accommodations It implies that Vietnamese people would like to enhance their sense of being ownership of a house This really plays an important role in improving the quality

of life for the citizens In other words, homeownership is not only helping people have the stable accommodation, but also helping the government be easier in administration and development local communities

In order to ensure its long-term sustainable development, the local government authorities have made better activities for housing sector by enforcing the law on residential housing to support homeownership for society, especially for low income families (Ho Chi Minh City Real Estate Association 2008) However, homeownership probability of a household always varies with changes in economy and society, especially

in the household characteristics such as household income, gender, marital status, education, age and household size (Hood 1999) In addition, homeownership IS

associated with financial ability (such as savings, credit) of a household (Mizuno &

Ishigami 1999) In addition to household characteristics, homeownership is also related to employment in which individuals would like to own a home if they can easily get jobs (Gabriel & Painter 2001) and immigration status due to its characteristics (Coulson 1999) because, nowadays, there are a lot of people moving to big cities from other provinces for living

It rs, therefore, interesting to find out what factors may influence the homeownership probability of every household in Vietnam This plays an important role for increasing of the opportunity of homeownership that is related to individual well-beings and development of the local communities (Cisneros 1995)

(ii) Socioeconomic characteristics: immigration status; credit accessibility and employment that implies the number of members working in a household

The thesis aims to establish a conceptual model for estimating the probability of homeownership on the basis of household and socioeconomic characteristics From results of estimating, policy recommendations are suggested

The result can be useful for policy makers They can base on the most significant determinants of homeownership to make positive policies for improving homeownership probability of households in Vietnam

1.3 Research questions

The study is to answer the following questions:

(i) How do household characteristics affect on homeownership probability of households in Vietnam?

(ii) How do socioeconomic characteristics affect on homeownership probability of households in Vietnam?

Specifically, the thesis aims to the following questions:

(i) Is the educational attainment level of the household head a mam factor of homeownership probability?

(ii) Does the employment really affect on homeownership probability?

(iii) Is the impact of immigration status significant on the homeownership probability?

1.4 Research methodology and scope of research

In order to have the right answers for research questions, econometric model and descriptive statistic methods are used to examine the relationship between the homeownership probability and its determinants

Specifically, the thesis is based on a Logit model for estimating determinants of the homeownership probability of households and a descriptive statistics analysis about the situation ofhomeownership in Vietnam

Data come from a secondary source, taken from Vietnam Household Living Standards Survey 2006 that is abbreviated to VHLSS 2006 This survey is carried out by the General Statistics Office of Vietnam, with the support of UNDP and World Bank Data is surveyed throughout the country As the thesis analyzes for the whole country, so the sample size will be 9,189 households throughout Vietnam extracted and regressed by STAT A

1.5 Structure of thesis

Following the introduction chapter, the thesis is continued with Chapter II in which the most relevant economic theories and empirical studies to homeownership and its determinants are reviewed An adaptation of economic theories and empirical studies to the homeownership model is analyzed Chapter III justifies methodology that includes analytical framework and research method In Chapter IV, an overview of homeownership is analyzed on the basic of VHLSS 2006 and housing sector development in Vietnam is presented Model regression by econometric analysis and evaluates the results, including presentation of data set, logistic regression results and results interpretation are introduced in Chapter V The last chapter is conclusions and policy implications

CHAPTER II

LITERATURE REVIEW

This chapter is developed to present a relevant theories and empirical studies of this empirical work and it is divided into two sections The first section is to review of theories that lay the theoretical foundation for this work It is organized into two parts The first part includes the definition of tetms and concepts related to the objects of the thesis such as home and homeownership In addition, house is recognized as durable goods that households, as units of analysis, would like to own The second part focuses

on the assessment of economic theoretical approaches that provides a review of economic theories on human capital investment, the decision making of a household on consumption, tenure choice decision and consumer theory for durable goods The assessment points out the roles not only of education, income and assets of the household

as their common resources, but also of household head's characteristics in the decision making of household's expenditures The decision making of the household on consumption by using their scare endowments to maximize their utility applied for homeownership Last, but not least, the macroeconomic variables of employment and immigration are introduced by the basic characteristic of the various approaches that the thesis deals with The second section analyzes empirical studies that can recognized more variables from the perspective of approaches, model and data used Moreover, the major findings from different empirical studies and the most significant variables on homeownership are introduced

2.1 Theoreticalliterature

2.1.1 Basic terms and concepts

The first is the definition of home and classification of home In the context of the thesis, home is the physical structure in which a household is living It could be a house

or an apartment But only house is researched because this is appropriate to Vietnam

Household Living Standard Survey 2006 by Vietnam General Office of Statistical that defines whether the household owns a house or not In reality, there are many different types of houses with different benefits in terms of homeownership However, some main types of houses that the thesis deals with are based on the definition of General Statistical Office (2007) in the questionnaire of Vietnam Household Living Standards Survey 2006, included "Villas"; "Strong houses with a private kitchen and bathroom/ toilet"; "Strong houses with a shared kitchen or bathroom/ toilet"; "Semi-permanent houses"; "Shift-made and other types of houses" These types of houses are designed for the survey by General Statistics Office that adapts to circumstances in Vietnam

The second is the definition of durable goods and consideration whether a house is durable goods According to Smith (1776), a house is a commodity, such as clothes and household furniture that is purchased for dwelling It is a consumption that does not make any profit or revenue but the expenditure for a house is really useful for the owner If a house is built well and maintained carefully, it may be used for many centuries

A commodity is considered as durable goods if it is expected to be used for three years and more such as electricity equipments, household fumiture, etc (McEachem 2006) Moreover, according to Household Living Standards Survey 2006, durables include houses, workshop, and etc So it is important to conclude that house is considered

as durable goods

The third IS the definition of homeownership and non-homeownership Homeownership concept is owner-occupied dwelling (Leppel 2007) in where a household is living In Vietnam, there are a lot of households own houses without any certificates issued by govemment In the context ofthesis, homeownership is not implied

by certificates issued by government, but it is based on the answer yes of question "Does your household own this residence?" (General Statistical Office 2007) Residence means one type of houses that is listed above A household that does not own a house as mentioned above is considered as non- homeownership That means they can rent a house

or live together with parents etc The homeownership that is only for dwelling does not make any revenue for its inhabitant but it is really useful to him In a case of a house for

rent, a tenant must pay the rent to the home owner That is to say, the homeownership can make revenue, though a house itself can produce nothing

The last is definition of household and household is considered as the unit of analysis for the study Himmelweit et al (1998) define: "household is a unit consisting of people with a common budget who organize some aspects of work and consumption together." This definition is appropriate to Vietnamese culture where all members live together in a family They together make decisions about the use of their budgets of human and material resources Another argument is that "individuals spend only their own income and spend it only on themselves But if income is shared in households, then the person who does the family shopping is neither spending just her own income nor buying just for herself." This is a reason for choosing a household rather than individual person as the rational decision making units in the theory of New Household Economics developed by Jacob Mincer and Gary Becker In summary, a household is recognized as the major consumption unit in many markets within a society (Schiffman et al 2001) In addition, the data of Vietnam Household Living Standards Survey 2006 is surveyed on the basis of a household that a group of persons usually live and eat together Therefore, household is chosen as the unit of analysis of the thesis

Furthermore, many recent household surveys are designed to ask questions to respondents in a household But nearly all of these questions are only asked to the household head The household head characteristics can represent for the overall subjective well-being of the whole household This may come from the reason that household heads live largely under the same conditions as other members in the household (Bookwalter et al 2006) So, most of variables in the model of the thesis are of the household head

2.1.2 Assessment of economic theoretical approaches

First of all, economic theories applied for the thesis are critically presented Many economic theories are used for studying of homeownership such as making the tenure decision that involves investment (Hood 1999; Leppel 2007), especially investment in education (Hood 1999) and consumption of a household (Leppel 2007) In addition, employment also affects on homeownership probability that is related to economic

performance (Sinning 2006) Furthermore, immigration characteristics (such as skills, wages and so on) are analyzed which these factors, in tum, influence homeownership ability of immigrants (Coulson 1999) Therefore, the relative areas of economic theories that mainly used for analytical framework apply the followings

(i) The human capital investment theory

The theory of human capital investment is very useful in analyzing the household decision on homeownership investment Basically, human capital investment in tenn of education attainment will influence on household eamings in the future and thus wi 11 influence homeownership decision Moreover, theoretical approach of investment in human capital is similar to that of investment in purchase of a house That is costs associated with the housing investment These costs are directly related to the purchase of

a house such as a down payment, other costs are paid every year during the ownership lifetime such as tax, maintenance fee and so on But there are also benefits associated with this investment Benefits of homeownership include such aspects as space (rooms and a yard) and more importantly, the pride of home owners Another important benefit

in the homeownership is the role of equity at the end of the ownership life That means at the end of his ownership life an individual has a major asset, namely a house which he may get capital gains from the investment (Hood 1999) A brief review of human capital theory is described as follow

According to Smith (1776), there are four types of fixed capital for a country, including (i) machines and trade's instruments; (ii) buildings; (iii) land; and (iv) people's abilities The first three types have considered as physical capital and the last one is just human capital that its characteristic is not transferable like others At the same time, an individual can also own these fixed capital

Nowadays, Becker (1993) has defined human capital concept as people's knowledge and skills which can be achieved by education and trainings Investment in human capital can make profit or influence on eamings because it is considered as

"physical means of production" that can make outputs increased In other words, investment in human capital increases the labor productivity This leads to an increase of income of employees The approach is based on the cost and benefit analysis of

investment in human capital (such as education at high school and college, on-the-job training for workers) The author focused on analysis of on-the-job training case that its general results can be used to apply for other types of human capital investments One characteristic of this kind of investment is that expenditure (cost) on education or training

is incurred at the beginning with the expectation for returns in the future Author shows that, in term of investment in human capital, education is the most important because it affects on a person's income People make expenditure on education in order to improve their knowledge This will lead to an increase in the eamings of educated people However, some specific investments in human capital do not have effects on earning to employees directly This is because workers cannot get returns from this investment that

is paid by firms In this case, firms will have gains directly But specific investments also benefit for workers because they create a long-term relationship between employers who pay costs of this investment and employees who become experienced in their jobs This leads to lower unemployment among skilled workers than unskilled ones That means skilled workers have stable jobs that stabilizes their annual incomes

Becker (1993) also asserts that the effect of human capital on earnings is different from person to person It depends on the age of individuals who invested in human capital According to age-earnings profiles, highly educated persons (college graduates and higher education) and less educated persons (high school or elementary school graduates) have the same characteristic of positive relationship between earnings and age But the average rate of increase in earnings is positive relationship to education This implies that college graduates have the higher average rate of increase in earning than high school graduates In summary, the investment in education will increase individual's mcomes

McConnell and Brue (1995) have applied microeconomics and macroeconomics theories to describe labor markets under the economic perspective Choice theory is used

in labor economics to explain behavior of individuals that takes part in the labor market One of assumptions of this approach is that individuals make choices today with an expectation of net gain in the future In the labor market, there is the difference in wage between skillful and unskillful workers Therefore, employees need to improve their labor quality This leads to invest in human capital in which education is a right choice

Consequently, educated persons earn higher income as compared to less educated people High educated persons also work hard because they think that it is a waste money and time of their investment if they do not work

It is important to say that, investment m education yields higher return to individuals, which in turn, he is able to purchase a house for accommodation

(ii) The decision making of a household on consumption, tenure choice decision in homeownership and consumer theory for durable goods

Before going further into the tenure choice of households in homeownership, the decision making of a household firstly assesses

Economics classifies expenditures into two types They are investment and consumption Investment expenditures make people expect an increased return in the future Whereas, consumption expenditures make people benefit in fom1 of utility immediately (Smith 2003)

Neoclassical economics identifies goods as normal, inferior and neutral goods As income increases, the demand for normal goods increases and the demand for inferior goods decreases For neutral goods, the demand is unchanged as income change (Begg et

al 2003) Since houses are durable goods They can be considered as normal goods So the demand for houses increases as incomes increase That is to say homeownership probability is likely to increases as household income increases

As mention above, a household is a unit of analysis in the thesis So "whenever economic units are not individual people, a question always arises as to how they make decision" (Himmelweit et al 2001) However, Samuelson (1956) considers household's decision making as a special case when a collective of members with different individual preferences make decisions as if they had a same preference A household utility function, accordingly, is the result of a consistent family consensus This represents a meeting of minds or a compromise between household members Author assumes the existence of a household welfare function to be maximized as a whole, with the level of household income given Simultaneously, it also maximizes the utility of all household members

Looking from a different perspective, Gary Becker, one of the founders of the New Household Economics, argued that a household consensus does not resolve the problem

of how different preferences among household members The author's analysis overcomes this problem of reconciling different preferences by showing how all members preferences of a household could be determined by the preferences of just one member, the household head (Himmelweit et al 2001) This means the household head's preferences effectively became the preferences of the household as a whole This theory indicates the importance of the head in the household's decision making The characteristics of the household head, therefore, are taken into account when examining determinants of homeownership

Furthermore, Himmelweit et al (2001) analyzes a "gender and power in Becker's model" Authors mention about the head of household who may be a man or a woman Normally, there are two parents with opposite sex and their children in the household Becker develops a model based on an implicit conception of gender relations in which author thought that a man is decision maker The reason is that "men were more likely to specialize in work market and thus contribute more income than home time to the household" This implies that men have a comparative advantage in the labor market and mainly contribute to households Other significant characteristics of men are "altruistic and redistribute towards other members of their families" That is why they have power

in decision making of households Contrary to men, women have the following characteristics: (1) A comparative advantage in the non- market work; (2) A less contribution to household; (3) A beneficiary of men's altruism; and ( 4) No power in decision making of the household These assumptions can be used for many criticisms If taking account of time as money, it is not clear that contributions of men are the most importance to households Because many women have full time jobs but they still do a lot

of domestic works as well as the feeding and care of children Thus it is worth to take the affect of gender in examining

In order to analyze decisions made within households, the New Household Economics assumes that "the household behave as if it had a single utility function." This implies that there are more benefits from living together than living alone between household members Consequently, there are generally gains from co-operative decision

within the household That is reason why people stay together in households It is to say that household size is important to be considered in the model concerned to households as the whole In addition, households have to make decisions that are related to the allocation not only of their members' labor but also of consumption of goods Households put resources (such as material, human, income, wealth) together in order to support people day-to- day living in the family and improve their welfare Authors also introduce two bargaining models of household decision making They are divorce threat model and separate sphere model These models focus on the decision making that would fare if the marriage break down and the couple divorce or the partners separately choose how much to contribute from their own gender-specific sphere This means the marital status has effects on household decision making

In summary, households include many individuals that have different sex and ages The power over decision making is not equal between households' members Most of the model of household decision making contain conception of gender that implies distinguishes men and women as well as their power within the household If considering the household as a family consensus or as the dictatorship of a household head, the household is a unit of analysis and household head characteristics should be concentrated the study on the probability ofhomeownership for households

The tenure decision involves both investment and consumption components In this thesis, it implies that households choose between owning and renting a house It deals with investment and consumption demand of the household If investment demand is greater than consumption demand then the household decides to own a home Otherwise,

if investment demand is lower than consumption demand then the household decides to rent a house (Leppel 2007) Clearly, households' decisions about housing are the result of

a behavior of investment or consumption on housing As an accommodation satisfies, consumption is need As a patrimonial asset, investment is considered (Arrondel &

Lefebvre 2001)

Consumer theory for durable goods

Durable goods play an important role in microeconomic analysis that involves durability choice of consumers The demand for durable goods has characteristics such as

(i) distinction ofpurchase (adding to stock) and consumption (physical deterioration); (ii) the purchasing decision based on the new information; (iii) Consumer confidence and income are the major determinants of purchases; (iv) the decision whether to buy or not is

a choice between two discrete alternatives (Deaton & Muellbauer 1980)

According to Glasner (1997), durable goods provide useful services to consumers for a long period of time Normally, purchase of durable goods has stronger fluctuation

on consumer spending than other consumer outlays With the viewpoint of spending on durable goods as the form of saving, the purchase of durable goods does not make decrease in net worth of consumers by the full amount of the purchase price The reason

is that the durable goods may become assets of consumers after buying Therefore, durable goods represents requirement of future consumption in which income may affect

on durables spending

Deaton and Muellbauer (1980) argued that consumer behavior in expenditure of durable goods depends on economic variables such as credit The credit restrictions are likely to affect through the stock adjustment process If consumers expect that the price

of durable goods will increase then they should buy now to make the big capital gain But consumers will not be taken this opportunity if they do not have enough liquid financial assets In addition, employment also has effect on demand for durable goods In the case

of high unemployment, durables purchases have been postponed

In summary, the demand for durable goods depends on preferences, assets, income and expectation in which every household is different Thus different households will have different behaviors in durables demand

(iii) Employment

Unemployment is one of major macroeconomic variables that reflect the performance of an economy as described in the book "Macroeconomics" written by Blanchard (2005) Author argues that if high unemployment happens, then (i) increases the probability of workers losing their jobs; (ii) reduces the probability of the unemployed finding a job and (iii) increases the duration ofunemployment

Figure 2.1: Wage and price setting and the natural level of employment

Source: Macroeconomics (Blanchard 2005)

As shown in Figure 2.1, employment N is measured on the horizontal axis The level of employment must be somewhere between zero and the labor force L For any employment level N, unemployment is U =L-N

As the level of employment increases, the real wage paid to workers increases So their incomes increase as well

Furthermore, there are a lot of studies that recogmze the negative impacts of unemployment on workers' well-being Employment provides not only material needs for the daily life but also various psychosocial benefits through functions such as (i) by imposing time structure on the working day; (ii) by implying regularly shared experiences and contacts with people outside the family; (iii) by connecting individuals to goals and purposes that transcend their own; and (iv) by defining aspects of personal status and identity, and 5) by enforcing activity (Jahoda, 1981)

Oliver and Shapiro (1995), in their analysis of both income and its translation into stable assets such as homes, cars and financial assets, show that work stability is very

important for the wealth accumulation capacities of people They also argue that work instability has a greater effect on increasing assets than it does on relative income But, in analyzing household in term of the socio- economic characteristics of the household like employment status, it is difficult to realize earnings of the household (Hsia & Chau 1978)

In summary, employment has a further positive effect on household income, which

m tum, to final demand of the household such as increasing opportunities to homeownership

(iv) Immigration

Finally, immigration characteristics that explain about immigrant's income, education and so on are described in the book "Immigration, Trade, and the Labor Market" edited by Abowd and Freeman (1991) is to analyze opportunities of homeownership for immigrants

In immigration studies, there are many approaches to research The first approach is the study of labor, goods and capital The second focuses on demographic that studies about the immigrant share of the labor force and the increase in the labor market The third one is related to trade that focuses on the balance of payments The last approach concerns on finance that considers the capital mobility (Abowd & Freeman 1991 ) In the context of thesis, immigrants' characteristics such as education, age, marital status, employment are focused

According to Becker (1993), investment in human capital is closely related to education and training However, McConnell and Brue (1995) have developed the concept of investment in human capital as migration A cost-benefit analysis approach is used to examine the migration A person will decide to migrate if the net present value of migration is greater than zero There are many factors that determine the migration such

as age, marital status, education, unemployment rate, and so on

- The migration of older people seldom occurs because it is difficult for the older individual to find a new job with which a large cost of investment in human capital is

compensated On the contrary, younger people are more likely to migrate because they would like to have a suitable job with their human capital investment

- Persons who get married are less likely to migrate because spouse may have a high salary job at living place In addition, if the whole family wants to migrate then the cost of migration may be higher than the benefit This leads to a negativity of the net present value of migration

- For education, the highly educated people are more likely to migrate because (i) they can get more opportunities for good jobs that make the higher gain from migration and (ii) the labor market for them is on the national range rather than local

- Households with more unemployed members are preferred to move out of their original location In addition, people will intend to migrate more if unemployment rate at their living place is high

With these characteristics, immigrants have lower homeownership rates than native households and that the homeownership gap widened significantly (B01jas 2002)

Conclusions of assessment of economic theoretical approaches

Homeownership is really important for households because of its beneftts Households have to make decisions that involve tenure choice between investment (owning of houses) and consumption (renting of houses) Various approaches have different assumptions but they are applicable to the study These are human capital theory, households' decision making and immigration approaches They all are related to gender, age-earnings profile, education, marriage and etc that is really appropriate for the study

2.2 Empirical studies

In this section, various empirical studies on homeownership are examined from the perspective of assessments about methodology and data used Simultaneously, findings and situations of empirical studies are essential assessments that might be using for the thesis

The dependent variable is the probability of homeownership that involves who owns a house and who does not own a house A household would prefer owning a house

than renting when considering benefits from homeownership in the long time (Hood 1999) However, it is not easy for a household to make decision between owning and renting a house because owners and renters have different behaviors that can lead to different benefits in economics (financial gain, assets accumulation) and use (satisfaction, enjoyment in accommodation) (Rohe & Stewart 1996) One of benefits in owner-occupation for the accommodation is the health People who care about health are now looking for owning houses increasingly (Easterlow et al 2000) In practice, the promotion of housing "choice" has implied the option to own to the most of households Then, people prefer to make the decision to own rather than to rent (Smith et al 2004) It

is, therefore, important to determine factors of homeownership

In the research "The determinants of home ownership: An application of the human capital investment theory to the home ownership decision", Hood (1999) has justified the human capital investment theory to apply for the homeownership based on cost-benefit analysis in which "a prospective home owner should invest in a house if the net present value of the benefits is greater than zero." Author argued that the model for homeownership is similar to the model of human capital investment These are costs related to investment in housing that owners have to pay for the first year such as down-payment, mortgage payments and insurance policy A few costs, such as tax, are paid every year during the ownership lifetime Addition to such costs, there are many benefits

of homeownership such as satisfaction of accommodations and the pride of the ownership Importantly, owner is likely to have house as a major asset So an approach of homeownership is designed that mainly divide factors into two groups: constraints and net benefits Constraints are race, gender and educational attainment The detem1inants of net benefits include age, marital status and household size As a result, a single equation logistic model to identify the relationship between homeownership probability and its determinants is:

ln[P/(1-P)] = Po + P1HHINC + P2WHITE + P3GEN + P4EDU + PsPARENT + P6AGE + P1MARITAL + PsHHSIZE + E

Where P is the probability of homeownership, HHINC is household income, WHITE is a dummy variable of ethnicity [white (1) and others (0)], GENis the gender of

respondents [male(1) and female(O)], EDU is education level, PARENT is the homeownership of parents, AGE is age of respondents, MARITAL is a dummy variable

of marital status [married group (1) and single group (0)], HHSIZE is the number of members in a household, are parameters, and 1.1 is error tenns

Data used for this research comes from the National Longitudinal Survey of Youth (NLSY) surveyed on the 1996 This data set is appropriate to the study because it includes data that is designed for collecting a respondent's demographic characteristics, income, wealth and housing choices

Hood (1999) finds that all independent variables, at a microeconomic aspect of homeownership, are positively significant except parental homeownership and large family size (from five people and larger) The reason is that as the number of people in a household increase, the costs within that household increase as well Thus, even with a high value of net benefits, a large number of households are dependent on a wealth constraint that does not allow them to invest in a house So author suggested a credit program to support households that have a lot of children in homeownership

Another research about dete1mining factors of homeownership associated with employment status and immigration used location choice and tenure choice models by Gabriel and Painter (2001) Authors estimate models by location and race to compare the homeownership differentials and the sensitivity of effects across location

Their study's approach is to examine the effects of household economic, educational, and other characteristics on both the direction of intra-metropolitan moves and the choice of tenure in the destination area on housing The main purpose of the research is to analyze the impact of household characteristics and preferences on location choice for homeownership The equation for location choice is a nested model It

includes location characteristics that vary across areas The impacts of household characteristics (such as age, gender, education, household size, number of workers in a household) on location choice are also included in the model Moreover, a probit specification is used to estimate determinants of housing tenure choice among each group

of mover households Thus, separate tenure choice equations are estimated for each group Immigration is added in this model Data used in this research are extracted from

the Public Use Micro-data Sample (PUMS) file of the 1990 decennial census that contains a five percent sample of all individuals living in LA County and San Bernardino County

Gabriel and Painter (2001) have found that increases in age of household head, married status, higher income, and higher level of education refer to increase in probability of homeownership An interesting finding is that a number of workers in the household have a "depressive effect" on the ability of home purchase This implies that, keeping constant of other factors, a household is less likely to own a house if additional workers are required to earn the same level of income In addition, immigrant population has a significantly negative effect on homeownership

In term of immigration, some empirical studies show evidences for an assimilation process between immigrant and native households in the context of homeownership In the paper of Sinning (2006): "Homeownership and Economic Perfonnance of Immigrants

in Germany", he [she] examines the homeownership gap between immigrant and native households in Germany Author used a double cohort approach to analyze the impact of the duration of residence in Gennany of immigrant households on the homeownership probability In order to examine the determinants of the homeownership probability, a binary Probit model is estimated by using the ownership status as a dependent variable Sinning (2006) identifies determinants ofhomeownership by the model:

P(Hh) =Po +Mh(Ch pl + Dh Pz + xh p3) + xh p4 + Ch, h = 1, ,N

Where P(Hh) is binary dependent variable of homeownership, Mh is a dummy variable for immigration [immigrant household (1), otherwise (0)], C11 is a vector of home country, Dh is the year of immigration, Xh refers to all independent variables that include household income, education, employment status, age, gender, marital status, household size, Pi is estimated coefficients and Eh is en·or with nmmal distribution In the model, there is an interaction between immigration and control variables

This research used data from the Getman Socio-Economic Panel (SOEP) The SOEP is a representative longitudinal survey that includes the sample of German and immigrant households Data is retrieved from 1984 to 2004 One of the major findings is that immigrant households have less opportunity to own a house than local households

Moreover, the findings indicate that the homeownership gap mainly happen m metropolitan areas as large concentration of immigrant households

Conclusions of empirical studies

All of these empirical studies have similarity in approaches and estimated models which are used to investigate determinants of homeownership Thus, such approaches and estimation models are selectively adopted for this study with critically considering the situation of homeownership in Vietnam Independent variables include not only household characteristics such as income, the number of household members, wealth, especially household head characteristics (such as education, age, marital status and gender), but also social-economic characteristics like credit, employment and immigration status The thesis is based on most of these factors at the micro and macro-economic level However, analysis in this work excludes ethnicity, location and some other variables Reasons are that it is because of legal, economic, political, and cultural institutions in which make homeownership study become more complex

Moreover, immigration status is also adopted for suitable for the thesis that implies

a moving from provinces to the big cities Regression methods of empirical studies are of

a single and multiple equations But a single regression equation is applied in this empirical work Empirical studies have provided a lot of significant variables On the basic of these empirical studies, variables are justified to be suitable to the situation of Vietnam that will be presented precisely in the next Chapter on research methodology

CHAPTER III

RESEARCH METHODOLOGY

This chapter is developed to aim at constructing the research methodology of the thesis This chapter is fallen into two sections Section 1 presents analytical framework which follows the research objectives, research questions and the literature review in the two previous chapters Section 2 is to describe the research method that is justified to be appropriate to the thesis in which an econometric model specification is drawn

3.1 Analytical framework

Based on the research questions and the literature review mentioned above, this section is subjected to introduce the analytical framework that should be explicitly presented in which methods of homeownership research are introduced with the major determinants of homeownership

3.1.1 Qualitative choice models for durable goods

As defined, durable goods can be used for a long-term, normally from three years

or more such as electric appliances, cars, and houses They are valuable goods That explains why the purchase of durable goods is implied to adding stock This is completely opposite to durables consumption which refers to depletion or physical deterioration (Deaton & Muellbauer 1980)

As a high value product, durable goods are not bought as frequently as consumable goods (such as clothes, rice, daily items) In expenditure studies, especially on durable goods, it is recognized that a large number of households do not have any purchase for a particular category during the survey period Consequently, the dependent variable will

be zero for a large number of observations In econometrics, this is referred as the limited dependent variable (Chern & Senauer 1993) It implies the choice is between ownership and non-ownership (discrete choice)

With considering x as income, £ as threshold of income level above which ownership P happens, so P = 0 if x < £ and P = 1 if x > £, Deaton and Muellbauer (1980)

showed that if we assume £ to be distributed independently of x and that it has the lognormal distribution with parameters f! and cr 2 g(x) is the probability of a randomly selected household owning the durable conditional on household expenditure x, then: g(x) = prob(P = 1; x) = f f(£; f!,cr 2)d£ =A( x; f!,cr 2)

Where A(x; f!,CY 2) is the cumulative distribution function of the lognormal distribution

3.1.2 Methods of homeownership research

The concept of owner-occupation in housing is different from country to country However, the tenure choice theory in housing is always an alternative for institutional explanations about homeownership probability There is a well-established framework which is developed for combining-'economics of tenure choice and housing demand Meanwhile, studies typically focus on supply considerations, often as a constraint It is because housing is held in short supply It is consistent with the active role of the government in supplying housing to its citizens In general, tenure choice studies in housing are associated with decisions to own or rent It is shown to be a function of the relative cost of owning versus renting, household demand for housing services, investment demand as well as credit and wealth constraints Some studies include tenure choice and mobility Most of the tenure choice literature examines only micro-level behavior, there is also an important considerations of institutional factors related to social, political, legal and cultural aspects of each society which may effect on the demand for homeownership (Fisher & Jaffe 2003)

In Quercia and Watcher (1996) work, authors propose a theoretical framework to determine abilities of a household in owning a home associated with tenure choice The household decision concerns about owning or renting a home Authors discuss factors of homeownership by classifying these variables into five groups: (i) assets, credit and income in which human capital (education) is often integrated in taking into account of permanent income, (ii) age, household size and family type are considered as life cycle of

a household, (iii) ethnicity which is normally used for examining the differences in income, asset, and so on for homeownership rate among ethnic groups of households, (iv)

house pnce and other market factors that influence the household investment and consumption, and (v) location and neighborhood characteristics associated with quality of life In addition, employment status is also included in their analytical framework

Household income and wealth are outstanding factors in determining the probability of homeownership in which wealth constraints are more important than income constraints (Linneman & Watcher 1989) Most of studies have found that income

is a statistically significant determinant of homeownership probability since households must have sufficient income for mmigage services that required by mortgage lenders The current wealth is a crucial factor in term of paying a down payment for a desired house Households will choose to rent if they are lack of the current wealth for purchase

of houses Besides income and wealth, credit history effects on loan approval As the result, it has affect on opportunities to own a house (Quercia & Watcher 1996)

In evaluating life cycle (such as age, gender and household size), Quercia and Watcher (1996) introduced three hypotheses that help to explain why life cycle status is also important in homeownership's ability First, different types of households have different demands for housing Second, some housing types are available for sale rather than for rent Finally, households that intend to move soon will prefer to rent In general, based on these hypotheses, households will make decisions on tenure choice that are better off for them Consequently, an alternative research design includes many important determinants to identify the probability of homeownership They are income, assets, household characteristics (such as age, gender, marital status and household size), employment status and credit However, the impact of immigration on homeownership has not paid attention

There is homeownership gap between different demographic groups This leads to a research on impact of immigration characteristics on the propensity for homeownership Normally, immigrant households have a low rate of homeownership There are some reasons for that First of all, immigrants need time to learn about the local real estate market or they intend to move again Second, they are not familiar with the local financial system that discourages the use of financial institutions for savings and debt to buy a house (Coulson 1999) Author used tenure choice as the dependent variable which

measures the probability of homeownership for each household It IS a function of determinants and a random: y = X + e

Determinants X includes income, education, marital status, location (sub-urban or central of city), householder's age and gender, immigration, nationality and the year that immigrants are entry the U.S

3.1.3 Major determinants of homeownership

On the basic of research questions, all important variables and the relationship between them are explained as follow

(i) Household income

Household income has both direct and indirect impacts on the probability of homeownership The direct influence is that when household income increases, the probability of owning a house increases Household would prefer to buy a house because

a higher income has more capacity to pay for initial costs Income has indirect relation to owning for a desired house because of the costs of homeownership Supposing the costs

of homeownership kept constant, as income rises, the proportion of those costs to income becomes smaller This creates more investment opportunities on homeownership for the high income households (Hood 1999) Besides initial costs, other costs of homeownership are mortgage payments, real estate taxes, and insurance that is called the user costs paid monthly The greater the user costs compared to renting, the less people is

to purchase a house (Leppel 2007) Goodman (1999) identifies tvvo types of income transitory and permanent income In the tenure choice decision, the permanent income has greater significant effect on the homeownership probability than the transitory income It is, therefore, income is a primary detem1inant of homeownership in which income from current income is important (Coulson 1999)

(ii) Household size

The number of members m a household will affect on the probability of homeownership The level of benefits of homeownership is increased in according with the more number of people in a household, especially the number of children An

increasing number of children in a household may lead to increasing in demand for homeownership However, large families (from five persons) with many children may often have financial constraints that can prevent homeownership Since the daily living expenses (such as food, day care, illnesses, etc) are the cost burden for the household with more children (Hood 1999) But Gyourko and Linneman (1997) found that there is

an increase in the homeownership probability for households with children compared to those without children The reason is that the presence of children in a family will create

a change in the discrete choice that is propensity to ownership

(iii) Assets

Besides income, Linneman and Watcher (1989) explored that wealth constraints have a negative impact on the probability ofhomeownership because it is very difficult to satisfy the mortgage qualifying requirements for housing purchase with less assets endowments This constraints the ability in borrowing Joan for individual homeownership propensity That is why Quercia and Watcher (1996) indicated that assets have concerned about financial capacity of a household to own a home A household can use wealth as collateral to borrow money for purchase a house The more cmTent assets that households have the more probability of homeownership increases

In addition to mortgage, assets are directly related to the higher down payment required for more costly houses It is not surprising that individuals with lower levels of educational attainment or declining their incomes often become homeowners due to better access to inheritance of assets from their parents (Gyourko & Linneman 1996)

(iv) Age

Age is considered as an important factor having influence on owmng a house because the elders often save more money to be able to cover in housing purchase and mobile less Homeownership's ability increases in common with age until retirement There are a number of reasons for this It may be a preference for ownership, the mobility decreasing, and the necessity of wealth accumulation (Coulson 1999) In addition, older households have higher income because they have spent more time in the workforce and get more experiences As a household's level of experience increases, the total income is more certainty in the near future Thus, older households are easier to commit to

homeownership They have time to be financially prepared for housing purchase Also, older households have more assets so that they can easily invest in housing This leads to increase in the probability of homeownership However, in the context of homeownership benefits, it is important to realize that there is a side effect As an individual grows older, his ownership life is shorter This makes the stream of benefits become shorter (Hood 1999)

(v) Marital status

According to Hood (1999), married people would prefer to increase the probability

of owning a home There are some reasons for that First, there is less mobility for married couples in comparison to single persons This leads to lower mmually transaction costs for accommodation and increase the ability of homeownership Second, married people can create more common wealth by pooling their incomes together By doing this, married people are able to overcome the wealth constraint that does not allow for homeownership as single persons Finally, married people often plan a future with children and they would like to have a stable environment to raise them Thus married people often consider making long term investment decisions with their common wealth

By considering homeownership as the asset and benefits that it provides, investment decision is the right choice Therefore, marital status is worth to consider as the determinant ofhomeownership

(vi) Gender

Another determinant of homeownership is gender Hood (1999) also argues that male often gets higher income than female who is often interrupted working for child bearing and rearing By working continuously over their work life without stopping working as female (due to childbirth) male have the oppmiunity to get more experience

in their jobs Therefore, male maybe have the more certainty of income than female Since male have these higher and more ce1iain incomes, they are able to secure a loan or mmigage Thus, male are more likely to become a home owner

(vii) Education

The level of educational attainment also impacts on homeownership High level of education may lead to a good job with a high salary This leads to increase homeownership ability because an individual with higher income will be able to serve the sufficient funds to cover costs occurred during the house purchase Moreover, well-educated people often save their income which creates necessary funds to secure a loan Thus, they have greater ability to be approved for mortgages on buying house Due to relationships between education, income and savings, an individual's educational attainment has effects on his homeownership decision (Hood 1999) Jn prospects of stability in the future, Feijten et al (2003) recognized that greater education level opens

up more job opportunities in which a comfortable working environment and the stability will be found This means a commitment in the life career are accepting As the result, well-educated people are more likely to become homeowners

(viii) Credit

In the recent studies, credit is chosen as a factor of detetmining types of homeownership whether people make decision to own or rent (Fisher & Jaffe 2003) In general, with the higher costs ofhomeownership, people need the credit assistant to cover down-payment and other costs paid monthly during the period of purchase However, credit creates barrier for individuals who want to become homeowners because of wage disparities and lack of collateral (Sinning 2006) Thus, the ability to get a credit becomes important to an individual to purchase a house It should be taken into account that lenders might impose requirements for its mortgage services that exclude most of the poor people (Ladd 1998) Therefore, Hood (1999) suggests credit assistance programs that increase opportunities of homeownership for low income and large families Author argues that a housing loan program is beneficial because of the higher costs of homeownership and the lack of wealth accumulation of the poor households

(ix) Immigration

Immigration status might be another determinant of homeownership Coulson (1999) claims the homeownership rate of immigrants is low since they might be young, less education and inaccessible to credit markets Immigrant experience may have adverse effects on the ability of obtaining information on housing market conditions or credit Due to unfamiliarity with financial institutions, immigrants may have a lower willingness to get the debt that is typically required for new homeowners This prevents them from becoming new homeowners In contrast, those immigrants who become citizens might be thought to be more stable and less mobile than others Thus they have incentives to absorb the large fixed costs ofhomeownership and a greater ability to obtain mortgage loans from financial institutions This leads to the willingness to become homeowners

In the prospects of economic performance, Sinning (2006) indicates that the ability

of immigrants to integrate successfully into the economic, social and political life of their new arrival places depends largely on their economic perfonnance In a certain place, the effect of the duration of residence on the homeownership probability has created a barrier

of ownership for immigrants, even though they can afford to buy a house Homeownership is considered as the long-tenn integration process of immigrants into society that plays an important role in providing long-term financial security Immigrants might face credit problems because the lack of wealth accumulation, income and employment As a result, the probability ofhomeownership for immigrants is low

(x) Employment

Gabriel and Painter (200 1) indicate that if a place has a Jot of job opportunities then people intend to own a house at that place In other words, job opportunities increase the probability ofhomeownership Authors also argue that "A household is less likely to own

if additional workers are required to eam the same level of income" while keeping other things constant This means that the number of workers (with a certain number) in a household have a diminishing effect on homeownership in this case

In addition, a steady employment might help the household in paymg their mortgage when there are many members in the household working For example, if the

wife and husband in the household have jobs, then the household have more income This leads to increasing accessibility to homeownership This effect is stronger when the young couples live together with their parents Also, the wife's employment in the labor market is an important factor when low-income households make the transition from renting to owning (Cabre & M6denes 2004)

3.2 Research method

3.2.1 Justification of models

According to Neoclassical economics, there are four models of consumer behaviors that focus on the analysis of the human's happiness as the main purpose They are revealed preference, marginal utility, indifferent curve and expected utility All of them are based on hypothesis in which all consumers (no differentiate between the rich and the poor) are constrained by their own budgets For marginal utility model, the logic is that when all marginal utilities per dollar are the same, the satisfactions must be maximized

In the indifferent curve model, with a given budget, the consumer will only choose one set of goods that his satisfaction is maximized Both of models indicate that budget and price of the commodity are major detem1inants of the demand for that commodity However, these models still have problems, especially with durable goods as noted below (Bumas 2002):

- With marginal utility model, consumers can get utilities beyond the current budget period for durable goods Alfred Marshall handles this problem by calculating the present value of the discounted flow of expected marginal utilities over the expected life of the goods

- Indifference curve model has the problem of binary demand For example, the household has to decide either buying a house or not If yes, the house can draw on one axis, the other axis for all other things If no, how to describe non house on one axis? In this case, the problem is the presentation of an indifference curve that is difficult to understand what its shape look like

One of the most crucial characteristics of durable goods is discrete choice This implies that consumers do not buy durable goods frequently, particular in continuous for