5609.0.55.003 - Information Paper: Changes to the method of estimating loan commitments to first home buyers , 2015
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 04/02/2015 First Issue
|Page tools: Print Page Print All|
FIRST HOME BUYERS
The ABS publishes monthly statistics relating to first home buyers in Housing Finance, Australia (cat. no. 5609.0). Loans to first home buyers are published in Table 9a (Australia) and Table 9b (State) on the downloads tab of the Housing Finance publication, and in Table 9 in the PDF. Tables 9 and 9a include the proportion of owner occupied loan commitments made to first home buyers.
First home buyer statistics are compiled using data from the ARF 392.0 Housing Finance form collected by the Australian Prudential Regulation Authority (APRA) under the Financial Sector (Collection of Data) Act 2001, and are an important input to analysis of housing demand and affordability. First home buyers are defined as persons entering the home ownership market as owner-occupiers for the first time.
An investigation by the ABS has identified that the data on first home buyers are under-reported, as some lenders have only reported loans to first home buyers who have also received a First Home Owner Grant.
The under-reporting only relates to loans to first home buyers. The total value of home lending is separately reported and is not affected.
The ABS and APRA are working with lenders to ensure all loans to first home buyers are recorded in the future, regardless of whether they receive a grant or not. In the interim, the ABS has adjusted first home buyer data for this under-reporting using estimates based on data provided by lenders which have reported correctly. The estimates will be updated over time as more lenders report correctly.
This Paper explains changes to the method of estimating first home buyers' data and the extent of revisions. The ABS will continue to update the adjustment method as lenders progressively improve their reporting.
The First Home Owner Grant (FHOG), introduced on 1 July 2000, is a national scheme funded and administered by the states and territories http://www.firsthome.gov.au. Under the scheme, a one-off grant is payable to eligible first home owners. Until October 2012, all first home buyers were eligible for the grant regardless of whether they bought a new or an established home.
Gradually, States and Territories restricted grants to new homes only so that first home buyers who were buying established homes were no longer eligible for the grant. APRA reporting instructions state that a First Home Buyer is a borrower entering the home ownership market for the first time as an owner-occupier. The instructions do not make any distinction between first home buyers who are eligible for a First Home Owner Grant and those who are not. Nonetheless, some lenders' reporting systems only record first home buyers if they are eligible for a grant which may cause under-reporting of first home buyers.
This under-reporting has progressively impacted on first home buyer statistics from October 2012 as individual States and Territories have changed the eligibility of their First Home Owner Grants, generally to cover only the purchase of newly constructed homes.
States and Territories restricted grants to new homes from different dates - New South Wales and Queensland from October 2012; Victoria from July 2013; the Australian Capital Territory from September 2013; South Australia and Tasmania from July 2014. Loans to first home buyers were therefore underestimated in these States from the dates specified due to some lenders under-reporting. Other lenders have reported correctly throughout. Originally, the drop in loans to first home buyers from October 2012 had been attributed to the change in grant eligibility reducing the affordability for first home buyers and economic conditions, such as rising house prices and the increase in investment loans for housing. However, subsequent analysis and follow-up with lenders has confirmed that the drop was due, at least in part, to under-reporting by some lenders.
CHANGES TO THE ESTIMATION METHOD
The ABS estimates that the number of loans to first home buyers which are currently being reported are approximately 80% of the total number of loans to first home buyers. Total reported monthly home loan commitments are not affected by this under-reporting.
For lenders who are under-reporting loans to first home buyers, the ABS has developed a model to adjust the proportion of first home buyers to total loans for each period of incorrect reporting. The model uses the following components:
b) the proportion of first home buyers to total loans for those lenders reporting incorrectly in the previous period;
c) the proportion of first home buyers to total loans for those lenders reporting correctly in the previous period; and
d) coefficients which determine the relative contribution of the above components to the incorrectly reported proportion.
The coefficients (d) of this model were estimated using data from January 2002 to the month prior to the First Home Owner Grant policy being changed (for example, in NSW the data were from January 2002 to September 2012). All the affected states were analysed separately. When more lenders are able to report correctly, the coefficients and estimates will be updated accordingly.
Application of the adjusted proportion:
The following table is an excerpt from the Housing Finance form (ARF392.0) and will be used to demonstrate the application of the adjusted proportion.
There are no known issues in reporting the total number and value of Fixed rate home loans (9t and 9vt), Secured revolving credit home loans (10t and 10vt) and Other home loans (11t and 11vt). The estimated proportion of first home buyers is applied to the totals for Question 9, Question 10 and Question 11 (9t, 9vt; 10t, 10vt; and 11t, 11vt) respectively to determine the number of first home buyers of the particular loan type. The values for non-first home buyers (i.e. All other loans) are then derived by subtracting the values for first home buyers from the respective totals.
Each lender reports the data by State and Territory, and the proportion for each period is applied to the relevant lenders at the state level. The adjustment is made at the lowest level collected, and is applied to the affected lenders and affected States only. The data are then aggregated to the published States and the national level.
Revisions have been made to the previously published data for the Number, Percentage (%) of all dwellings financed, and Average loan size of First Home Buyers and Non-first home buyers at the national level (columns B to G of Table 560909a). Relevant States' previously published data have also been revised (Table 560909b) back to when the First Home Owner Grant was first restricted in that State or Territory.
For more information, please refer to the Appendix. If you have any feedback on this Paper, please email email@example.com or phone the Financial Statistics Section on Sydney (02) 9268 4186.
The graphs below illustrate the impact of the adjustments to previously published estimates of the number of loans to first home buyers.
Adjustments affect all States and Territory estimates of first home buyers other than Western Australia and Northern Territory. Should under-reporting occur in Northern Territory when grant eligibility changes from January 2015, the same adjustment method will be applied. Adjustments to estimates in Western Australia are not expected because a first home owner in this State is still eligible for the grant if they buy an established or a new home. Illustrative graphs for Australia, New South Wales, Queensland, Victoria and Australian Capital Territory are provided below. Graphs for South Australia and Tasmania are not included as changes to their First Home Owner grants have only been introduced recently. The same method of adjustment has been applied to these two States.
These documents will be presented in a new window.