Housing equity and household consumption in retirement: evidence from the Singapore Life Panel^©

ABSTRACT

Housing affordability for elderly homeowners involves an entirely different set of issues as compared to housing affordability for first-time homeowners. To afford to ‘age-in-place’ may require homeowners to access channels that enable them to withdraw their housing equity to finance consumption in retirement. We utilize data from the Singapore Life Panel^© survey to empirically investigate the impact of housing equity on the consumption of elderly households. Based on panel analysis, we find housing equity value has no significant impact on non-durable consumption for elderly people. The conclusion holds for a battery of robustness checks. Moreover, heterogeneity analyses based on subsamples by the health condition, the age of household head, the house type, and the number of properties owned also show no significant impact of housing equity on consumption. Finally, we use scenario analysis to study the Lease Buyback Scheme (LBS), a novel housing equity monetization scheme that allows elderly households to unlock housing equity for retirement financing. An individual scenario analysis reveals positive but negligible effects, which may explain the low take-up rate for the LBS.

KEYWORDS:

• Housing wealth
• elderly households
• housing equity withdrawal
• Singapore

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

Source: Data from Singapore government websites.

Source: Housing and Development Board website at https://www.hdb.gov.sg/cs/infoweb/residential/living-in-an-hdb-flat/for-our-seniors/monetising-your-flat-for-retirement/lease-buyback-scheme

Notes: This table reports the summary statistics of the whole sample. Panel A shows the statistics of dependent variables. Panel B shows the statistics of explanatory variables. Panel C shows the statistics of control variables, including age and marital status.

Notes: This table reports the estimates of Model (1). Dependent variables and explanatory variables are log transformed. All specifications include control variables (age and marital status), family fixed effects, and wave fixed effects. The t-statistics are provided in parentheses below the estimates. Standard errors are clustered at the household level. * denotes p < 0.1, ** denotes p < 0.05, *** denotes p < 0.01.

Notes: This table reports the estimates of Model (1) using subsamples based on the health condition of household members. Dependent variables and explanatory variables are log transformed. Columns 1 and 3 restrict the sample to households with at least a household member with chronic disease(s). Columns 2 and 4 restrict the sample to households whose average monthly health expenditure exceeds the 75th percentile of the health expenditure distribution. The t-statistics are provided in parentheses below the estimates. Standard errors are clustered at the household level. * denotes p < 0.1, ** denotes p < 0.05, *** denotes p < 0.01.

Notes: This table reports the estimates of Model (1) using subsamples based on households’ house type. Dependent variable and explanatory variables are log transformed. Column 1 restricts the sample to HDB households. Column 2 restricts the sample to non-HDB households. The t-statistics are provided in parentheses below the estimates. Standard errors are clustered at the household level. * denotes p < 0.1, ** denotes p < 0.05, *** denotes p < 0.01.

Notes: This table reports the estimates of Model (1) using subsamples based on the age group. Dependent variable and explanatory variables are log transformed. Column 1 restricts the sample to households with a household head aged between 50 and 63. Column 2 restricts the sample to households with a household head aged between 64 and 70. The t-statistics are provided in parentheses below the estimates. Standard errors are clustered at the household level. * denotes p < 0.1, ** denotes p < 0.05, *** denotes p < 0.01.

Notes: This table reports the estimates of Model (1) using subsamples based on the number of property owned by households. Dependent variable and explanatory variables are log transformed. Column 1 restricts the sample to households with one property. Column 2 restricts the sample to households with more than one property. The t-statistics are provided in parentheses below the estimates. Standard errors are clustered at the household level. * denotes p < 0.1, ** denotes p < 0.05, *** denotes p < 0.01.

Notes: This table reports the estimates of Model (3) in the scenario analysis in Section 4.4. Dependent variable is log transformed. The t-statistics are provided in parentheses below the estimates. Standard errors are clustered at the household level. * denotes p < 0.1, ** denotes p < 0.05, *** denotes p < 0.01.

1 Vaithianathan et al. (2018) compare monthly household income and expenditure recorded in the SLP with those in official statistics, and they find that the data is very similar in distribution.

2 All references to $ in this paper are to Singapore dollars. The exchange rate on 14 October 2020 was S$1.36 to US$1.

3 Comprising 830 households in four-room HDB flats, 2,030 households in three-room HDB flats, and the remaining 240 households in smaller HDB flats. (As revealed in Parliament by the Minister for National Development, and reported by Channel News Asia, Oct 1, 2018). Access at: https://www.channelnewsasia.com/news/singapore/3-100-households-have-taken-up-lease-buyback-scheme-mostly-3-10775596.

4 Access on August 12, 2020. https://www.hdb.gov.sg/cs/infoweb/residential/living-in-an-hdb-flat/for-our-seniors/monetising-your-flat-for-retirement/lease-buyback-scheme/how-it-works.

Additional information

Funding

Liang Jiang’s research was supported by MOE (Ministry of Education in China) Project of Humanities and Social Sciences (Project No.18YJC790063). Sock-Yong Phang’s and Jun Yu’s research was supported by The Ngee Ann Kongsi, and by the Ministry of Education, Singapore, under its Academic Research Fund Tier 3 programme (Award reference numbers MOE2013-T3-1-009 and MOE2019-T3-1-006).