FREE ELECTRONIC LIBRARY - Theses, dissertations, documentation

Pages:     | 1 |   ...   | 69 | 70 || 72 | 73 |   ...   | 156 |

«VOLUM E 1 1, N UM B E R 1 I S SN 2 1 6 8 - 0 6 1 2 F L ASH DR I V E I S SN 1 9 4 1 - 9 5 8 9 ON L I N E T h e In s t it ut e f o r Bu s i n e s s an ...»

-- [ Page 71 ] --

Hu (2004) analyzes whether owning a house affects the ratio of risky asset investments and finds that renters are apt to be more risk averse in allocating portfolio assets. Therefore, he provides evidence that the share of stocks to liquid assets for house owners is higher than the share of stock to liquid assets for renters. Yao and Zhang (2005) extend their interest on housing and portfolio choices of liquid assets between stocks and bonds. In the home owner’ investment decision to participate in the stock market, the share of the house price to net assets negatively affects the household portfolio and the share of the house mortgage debt to net assets positively affects the household portfolio. Cocco (2005) shows that investment in a house asset plays a role in explaining the cross-sectional variations of asset components and stock holdings, thereby analyzing the effect of the risk in house assets on the portfolio of financial assets. He provides evidence of the effect of crowding out households that exclude risky assets, such as stocks in the formation of a portfolio of financial assets, when home owners face risks of decreasing house prices. In a recent study by Chetty and Szeidl (2014), they estimate a model with separate effects for mortgage debt and house assets on household portfolios. To address the endogeneity problem, they use the average house price and house GCBF ♦ Vol. 11 ♦ No. 1 ♦ 2016 ♦ ISSN 1941-9589 ONLINE & ISSN 2168-0612 USB Flash Drive 322 Global Conference on Business and Finance Proceedings ♦ Volume 11 ♦ Number 1 supply elasticity as instrument variables and generate estimates with separate exogenous variations. Using PSID of the U.S., they find that an increase in mortgage debt significantly affects a decrease in stock holdings, but an increase in house value significantly affects an increase in risky assets. Therefore, they provide evidence that the effect of housing on portfolio choices exists.


Our study uses SHFLC for the years 2010-2013, conducted by the Korea National Statistics Office (NSO).

Since the initial sample selection, a new survey has been started. The sample now constitutes two different waves: 2010-2011 and 2012-2013. Currently, the NSO is continuously surveying the same households from the 2012 sample. The data contain demographics, such as age, gender, education level for the head of the household, and main occupation type for each household, as well as broad financial information, for example, mortgage debt, other financial debt, and asset estimations. In our sample, since we are not interested in whether or not individuals own their houses, we focus only on households with house ownership. Our starting sample is 22,952 households. Sample selection deletes the following: i) head of the household age below 20 or over 80, ii) financial assets less than zero, and iii) negative housing equity.

Therefore, our final sample is 22,221 households.

Our empirical model is based on the model suggested by Chetty and Szeidl (2014). To analyze the effect of housing ownership on the household portfolio through exogenous variations of housing assets and mortgage debt, they derive an analytical approximate equation for optimal portfolio choices under a typical (stylized) two-period model. Their suggested model has the advantage of providing a simple, easy, and tractable tool, combined with insight into other important mechanisms affecting the household portfolio choices including illiquidity and housing price risk, hedging effects, and variation effects. In addition, considering fixed moving costs, multi-period, and labor income risk, the model can be generally characterized. The empirical model in its explicit form derived from Chetty and Szeidl (2014), analyzes the separate effect of the housing asset and housing mortgage debt on a household’s portfolio. The model

follows a linear combination:

, = 0 + 1 , + 2 ℎ, + , + ,. (1) In equation (1), , is the dependent variable for each household i in year t; , is mortgage debt (or property value); and, ℎ, denotes the housing asset. , is the vector of the control variables, and these are divided largely into two groups. The first group is the demographic variables: household head age, education, gender, and marital status, household size, a dummy for the metropolitan area location; and, the other variables are yearly dummies. For the benchmark model, we extend two models with different results are 1 0 and 2 0. An increase in mortgage debt decreases the share of stock holdings through dependent variables, using two different risky assets, which we explain in the data. The expected empirical the general increase in illiquidity, an increase in risk exposure, and a decrease in the present value of life assets. Simultaneously, an increase in the housing asset affects the share of stocks to liquid assets through the increased wealth effect, and the expected positive effect (Yao and Zhang 2005).


The Effect of Housing on the Household Portfolio As explained in the model, we expect the empirical results to be 1 0 and 2 0. However, the residual 1 could have other sources of heterogeneity when a household is making portfolio choices. As pointed out by other studies, these different kinds of sources are entrepreneur risk, investment failure, risk-aversion heterogeneity, and income measurement error. There exists a correlation between mortgage debt and the

–  –  –

estimating 1 and 2. In our study, first we compare our Ordinary Least Square (OLS) results with the housing asset and the residuals caused by heterogeneity; clearly, if we use OLS, then there is bias in empirical results of other existing studies, and then to separate the independent effect of mortgage debt and the housing asset on the portfolio, we utilize the estimate approach by Chetty and Szeild (2014), using Korean household data.

The results using OLS for equation (1) are shown in Table 1. The empirical results are from four different models, which are combinations with different explanatory variables. First, for the two major points of The first model is without control variables, and the coefficient of mortgage debt 1 is positive, and the interest, with mortgage debt and the housing asset, we analyze combinations with other control variables.

coefficient of house asset 2 is significantly positive. When we control the demographic variables, 1 changes to significantly negative, while 2 is still significantly positive. In the fourth column of the Table, the model with all the control variables, the coefficient of mortgage debt 1 is significantly negative, and the coefficient of the housing asset 2 is significantly positive.

These results imply the importance of controlling for the demographics. However, it may not make a big difference to the empirical results considering the fixed effect from the short panel of two years. Our results are consistent with the prediction of the theory, even though we use the OLS estimation method. Our results, with OLS, are in contrast to the results from existing research of the positive correlation between mortgage debt and risky asset portfolios (Heaton and Lucas, 2000, Cocco, 2005, Yao and Zhang, 2005, Chetty and Szeild, 2014). For the robustness of our empirical results using a relatively low share of stocks to liquid assets, we extend the empirical test using two alternative dependent variables. One dependent variable, risky1 is defined as stocks added to future options, and the other variable, risky2, adds deposit savings.

Deposit savings is the concept of a lower risk asset, and if we include this, then the share of risky assets to 2 and 3. In Table 2 in the fourth column, 1 is significantly negative and the coefficient of the house asset liquid assets rises above 20%. The results with concept for the extended risky assets are described in Tables 2 is significantly positive, with different magnitudes of its coefficient compared to Table 1. The results for all the models are similar to the results in Table 1, even though we include a wider definition of the dependent variable. The only difference is that the magnitude of the coefficient is larger since the mean of the dependent variable is larger than our initial dependent variable.

The Effect of Housing on the Portfolio Using Instrument Variables

As in previous research, we use instrument variables to solve the endogeneity problem. We utilize a research approach for this suggested by Chetty and Szeild (2014). The basic idea of the approach is that we derive general variations of mortgage debt and housing assets using two house prices as instrument variables. Two different house prices are considered: the average price of a house in the household’s state in the current year, and the average price of a house in the household’s state in the year that the house was purchased.

However, our study has two limitations compared to the research of Chetty and Szeild (2014)`s for U.S, households. First, our data lack information about the purchasing date by the current household. Therefore, we use only current house prices in the household’s area in the current year. Second, the ranges of crosssectional and time-series house prices are shorter than the state-level and the time-series in the U.S. To make up for such gaps, we try to maximize our explanatory power related to cross-sectional variations of the instrument variables.

GCBF ♦ Vol. 11 ♦ No. 1 ♦ 2016 ♦ ISSN 1941-9589 ONLINE & ISSN 2168-0612 USB Flash Drive 324 Global Conference on Business and Finance Proceedings ♦ Volume 11 ♦ Number 1 Table 1: Empirical results: dependent variable – stock share (%)

–  –  –

Therefore, we use house prices for house types as well as areas in which the household resides in the current year. Our house residency type is classified as detached house, apartment, row house, and others. The Korea National Statistics Office (NSO) provides a yearly house price index for each house type and we use these house price indices to match each household’s house type. Additionally, we collect information for two different areas: the metropolitan and non-metropolitan areas. In the same way, we match the house price index for the metropolitan and non-metropolitan areas. The results are shown in Table 4. We use a twostage least square (2SLS) analysis and we first regress the instrument variables on the dependent variable, where the explanatory variable has endogeneity; in the second stage, we use the prediction value from the first stage as the explanatory variable. In this study, we analyze the average house price as the instrument variable for mortgage debt and the house asset.

The first column (i) of Table 4 presents the result for the first-stage of the mortgage debt, with both yearly dummies and demographics. The second column (ii) provides results for the dependent variable as the house asset in the first-stage. The estimation result shows a strong positive correlation between each mortgage debt and house asset and house price index. In the second stage, we estimate the explanatory variable in equation (1) as the predicted value using the instrument variable in the first-stage. The coefficient of this is consistent with theory prediction results. The negative estimate of 1 implies a negative (1.70)% mortgage debt is significantly positive and the coefficient of the housing asset is significantly negative, thus estimate of 2 implies a positive 1.96% increase in the stock-holding ratio as the house asset increases by stock-holding ratio drop as mortgage debt increases by KRW 1,000M and simultaneously a positive KRW 1000M.

–  –  –

Number of obs. 21,980 This table shows the effect of mortgage debt and housing assets on the share of stock in a household`s portfolio using 2SLS. we use the average house price in the current year as the instrument variable for mortgage debt and the house asset. for controlling household`s demographic factors and year dummies.


In this paper, we analyze the correlation between housing and the household portfolio with a cross-sectional instrument variable using data from The Korean Household Finance and Welfare Survey for the years 2010Deriving an empirical model from a theoretical one, we use the instrument variable to solve the endogeneity problem where the explanatory variable and residual are correlated. To this end, we use the average house price in the current year as the instrument variable for mortgage debt and the house asset.

House price indices are used as various variations for metropolitan and non-metropolitan areas and house type. We found that the effect of housing on the household’s portfolio is significant. An increase in mortgage debt has a significantly negative effect on the share of stocks to liquid assets and an increase in coefficient 1 implies a negative (1.70%) stock-holding ratio drop as mortgage debt increases by KRW the house asset has a positively significant effect on the share of stocks to liquid assets. A negative 1,000M and simultaneously a positive coefficient 2 implies a positive 1.96% increase in the stock-holding ratio as the house asset value increases by KRW 1000M. We expect that related research can expand in the future using this research as a starting point.


Campbell, J. Y. and Cocco, J. F. (2003), ‘Household Risk Management and Optimal Mortgage Choice,’ Quarterly Journal of Economics, 118, 1449-1494.

–  –  –

Chetty, R. and Szeidl, A. (2014), ‘The Effect of Housing on Portfolio Choice,’ Working Paper.

Pages:     | 1 |   ...   | 69 | 70 || 72 | 73 |   ...   | 156 |

Similar works:

«General Certificate of Education Advanced Level Examination June 2012 Drama and Theatre Studies DRAM3 Unit 3 Further Prescribed Plays including Pre-Twentieth Century Thursday 31 May 2012 1.30 pm to 3.30 pm For this paper you must have: an AQA 16-page answer book  the texts of the set plays you have chosen.  Time allowed  2 hours Instructions  Use black ink or black ball-point pen. Use pencil only for sketches and diagrams.  Write the information required on the front of your...»

«European Wind Energy Conference & Exhibition 2008 Press Clippings TABLE OF CONTENT TVs: – Euronews – GreenTV (UK), 250,000 visitors per month – EurActiv (along with the article) (requests session) Newsmarket (requests for video of the opening session): – 24horaswallstreet.tv – Above the Line Programming – ARD / SWR – Artenergy Publishing – ATN Bangla UK (Sky-827) – Bloomberg – Blue Man Publishing – codesign – DAF (Deutsches AnlegerFernsehen) – Delphic Insight –...»

«Trinity College Dublin Life within the microcosm: 100 years of Trinity Hall In partial fulfilment of the M. Phil degree in Gender and Women’s Studies Author: Rosa Pilcher Supervisor: Maryann Valiulis Date submitted: 11th September 2009 Abstract: In 2008, Trinity Hall in Dartry Road proudly celebrated its Centenary. The student residence, which now accommodates somewhere in the region of 1000 male and female students, was originally created in 1908 as a place to house the new female students...»

«LCB File No. T053-99 ADOPTED TEMPORARY REGULATION OF THE NEVADA TAX COMMISSION (Effective June 11, 1999) EXPLANATION – Matter in italics is new; matter in brackets [ ] is material to be omitted. AUTHORITY: NRS 360.090, NRS 372.025, NRS 372.050, NRS 372.055, NRS 372.085, NRS 372.725 AND NRS 372.735.Section 1. NAC 372.460 is hereby amended to read as follows: Sec. 2. 1. The sale of tangible personal property by a retailer includes the furnishing [, pursuant to the guaranty provisions of the...»

«EN EL TRIBUNAL SUPREMO DE PUERTO RICO Olga Domínguez Castro Sandra J. Guzmán Hernández, Militza López Mateo, Carlos Rivera Figueroa Recurridos Certiorari v. 2010 TSPR 11 Gobierno del Estado Libre Asociado de Puerto Rico, Secretario de Justicia, 177 DPR Honorable Luis Fortuño, Gobernador del Estado Libre Asociado de Puerto Rico, Departamento de la Familia y Junta de Reestructuración y Estabilización Fiscal Peticionarios Número del Caso: CT-2009-4 CT-2009-5 CT-2009-6 CT-2009-9 Fecha: 2...»

«Research Scholar ISSN 2320 – 6101 www.researchscholar.co.in An International Refereed e-Journal of Literary Explorations Impact Factor 0.793 (IIFS) FAMILIAL METAPHOR IN THE POEMS OF A.K. RAMANUJAN WITH SPEACIAL REFERENCE TO OF MOTHER AMONG OTHER THINGS & OBITUARY Dr. Richa Tiwari Guest Faculty Department of English Swami Vivekanand Govt. College, Lakhnadon, Seoni (M.P.) Ramanujans poetry emanates from his desire to come to term with himself against the background of expatriation and...»

«MDRC Working Papers on Research Methodology A Conceptual Framework for Studying the Sources of Variation in Program Effects Michael J. Weiss Howard S. Bloom Thomas Brock June 2013 Acknowledgments The ideas, writing, editing, and review of this paper involved a great many people who vastly improved its quality. In particular, we would like to acknowledge the contributions of MDRC staff members Gordon Berlin, Ginger Knox, Shira Mattera, James Riccio, and Marie-Andrée Somers, and consultant Kay...»

«MICHAEL KOHLHAAS HEINRICH VON KLEIST En las riberas del Havel vivía hacia mediados del siglo XVI un tratante de caballos llamado Michael Kohlhaas, hijo de un maestro de primeras letras, y uno de los hombres más honrados y a la vez más terribles de toda su época. Hasta sus treinta años de edad, este hombre tan fuera de lo común hubiera podido ser considerado como modelo de ciudadanos. En una aldea que todavía hoy lleva su nombre, poseía una granja en la cual vivía tranquilamente con lo...»

«Action Plan A recommendation to Governor Kulongoski from the Ending Homelessness Advisory Council June 2008 Ending Homelessness Advisory Council CHAIR Rick Crager, Deputy Director Thomas McClellan, Administrator Oregon Housing and Community Services DMV, Oregon Department of Transportation VICE CHAIR Mitch Morrow, Deputy Director Commissioner Bill Hall Oregon Department of Corrections Lincoln County Commissioner Tom Olsen, Child Care Administrator Edward Angeletti, Planning Manager Oregon...»

«Second-Level Digital Divide: Differences in People’s Online Skills* Eszter Hargittai** http://www.eszter.com Sociology Department Princeton University This paper is a pre-print copy of the article that appeared in First Monday available online at http://www.firstmonday.org/issues/issue7_4/hargittai Abstract Much of the existing literature on the digital divide – the differences between the haves and have nots regarding access to the Internet – limits its scope to a binary classification...»

«How To Raise A Boyfriend A has they can actually How to Raise a Boyfriend commute a schedule terms and put a income focusing about your website course outsource. When your systems hinder helping off, also the programs you may expect to fix your settlement. Be additional to come about the glossy websites, and have in your debts are removed and purchased. Also I generally need to get your next key in usually, but take out to walking the anything for duration goals if it ca be down. A resource...»

«Three Oswego County Brothers in the Civil War and After: A Review of the Scriber Family and the 24th, 110th, and 184th NYSV Infantries By Kent Scriber Ithaca, New York Three Oswego County Brothers in the Civil War and After: A Review of the Scriber Family and the 24th, 110th, and 184th NYSV Infantries by Kent Scriber Ithaca, New York January 3, 2014 Danby Press Ithaca, New York Contents Chapter 1 Introduction..7 Chapter 2 Early Background.. 13 Chapter 3 Oswego County Answering the Call..19...»

<<  HOME   |    CONTACTS
2016 www.theses.xlibx.info - Theses, dissertations, documentation

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.