Understanding New Zealand’s decline in homeownership

Abstract

Homeownership is an important component of the New Zealand lifestyle. In recent decades, however, the ownership rate has been declining and the reasons are poorly understood. This paper explains the decline using a decomposition technique that has been applied in other contexts. We find that borrowing constraints and ethnicity have been particularly important contributors to the decline. Rapidly rising house prices clearly have played a major role in the inability of income to keep up with prices and the increased impact of borrowing constraints. We also show that the increased down payment requirements imposed by the Reserve Bank of New Zealand in 2013 are unlikely to have affected the ownership rate.

Keywords:

• Homeownership
• housing finance
• New Zealand

JEL classifications:

• G21
• R21

Acknowledgement

We are grateful to David Tripe for helpful comments and to the staff of the Statistics New Zealand Data Lab for assistance. Access to the Household Economic Survey data used in this study was provided by Statistics New Zealand under conditions designed to give effect to the security and confidentiality provisions of the Statistics Act 1975. The results presented in this study are the work of the authors, not Statistics New Zealand.

Notes

1. A comparison of homeownership rates across 24 developed countries puts New Zealand at about the median based on the 2006 census rate of 66.9 per cent (Bourassa et al., 2013). Ownership rates in these countries varied from a high of 87.2 per cent for Singapore to a low of 34.6 per cent for Switzerland.

2. Similar views are also presented in a report prepared for Commission for Financial Literacy and Retirement Income by Saville-Smith (2013) and in other countries, such as Australia, where homeownership has served as an important source of financial support during retirement (Yates & Bradbury, 2010).

3. However, there is no capital gains tax for family homes or for residential investment properties.

4. Under the KiwiSaver scheme, eligible first home buyers are able to buy a home with their savings in the KiwiSaver, plus a maximum $5000 subsidy from the government for the deposit. Under the Welcome Home Loan scheme, first home buyers pay a low deposit (currently set at 10 per cent of the purchase price) and Housing New Zealand (a state-owned enterprise) underwrites the loan for private lenders. The assistance is subject to house price caps and income limits.

5. Auckland house prices increased by 14 per cent in 2013 and 10 per cent in 2014.

6. This market-oriented approach to the rental sector did not change during or after our period of study, nor are any changes anticipated.

7. RBNZ data table C22 shows that housing and land comprised in excess of 50 per cent of household wealth as of December 2014.

8. The optimal house value is based on a regression of house value (for current owners) on an instrument for salary income, user cost and location dummy variables.

9. Further details about these instrumental variables can be found in Bourassa (2000).

10. The hedonic equations specify house value or annual rent as a function of the numbers of bedrooms, bathrooms, living rooms and other rooms, a dummy variable for detached homes (attached is the default category) and dummy variables for regions (Wellington, Christchurch or the rest of New Zealand, with Auckland being the default category). Additional details for the hedonic estimations are available from the first author.

11. For further details about measurement of the user cost in New Zealand, see Bourassa (2000). We provide a relatively brief overview of this here because the relative cost ratio is not significant in our estimations and was dropped from the final analyses. One interpretation of these results is that the culture of homeownership in New Zealand is such that households buy whenever they can afford to without regard to the relative cost of owning and renting. This underscores the important role that homeownership plays in New Zealand.

12. Two other demographic characteristics were tested in initial estimations but omitted from the final estimations because they were never statistically significant: the number of persons in the household and the gender of the reference person.

13. For discussion of the calculation of marginal effects for logit models, see Greene (2007), section R10.4.

14. In New Zealand, these values are considered to be reasonably current and reliable indicators of market values (Shi et al., 2009). Under the Rating Valuation Act 1998, all residential properties in New Zealand are required to be reassessed to market levels on a regular basis. This is typically done by the local authorities every three years. Statistical studies commissioned by the Office of Valuer General in New Zealand show that the overall measurement errors in assessed values are small throughout the housing stock and well within the International Association of Assessing Officers statistical standards.

15. In their survey of city apartment markets in New Zealand, Hargreaves & Shi (2005) found that the annual property maintenance and repair cost including rates and insurance but excluding any property depreciation was on average 2.3 per cent in Auckland and 2.1 per cent in Wellington. Property depreciation is typically assumed at 1 per cent per annum.

16. The negative user costs and relative cost ratios are consistent with the findings of Shi et al. (2014).

17. Although the relative cost ratio is no longer in the model, expected house price inflation still plays a role via the user cost variable in the optimal house value equation.

18. Note that the rankings of the behavioural effects in Table 4 exclude the impacts of the constant terms, which can be interpreted to capture the effects of unmeasured variables.

19. This implies that the bootstrapped confidence interval for the simulated ownership rate is irrelevant.