5609.0 - Housing Finance, Australia, August 2017 Quality Declaration
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 12/10/2017
|Page tools: Print Page Print All|
3 All lending commitments are classified to the lender type which is (or will be) the legal lender on the corresponding loan contract. Commitments are published for two broad groupings of lender type, Banks and Non-Banks. The Non-Bank grouping also has the components Permanent Building Societies and Wholesale Lenders n.e.c. published.
4 Housing loan outstandings are classified to the following lender types: Banks; Permanent Building societies; Credit unions/cooperative credit societies; Securitisation vehicles; and Other lenders n.e.c.. The first three of these types are components of the grouping Authorised Deposit-taking Institutions (ADIs). Loan outstandings for the ADI lender types are published monthly, and are classified by purpose (owner occupied housing or investment housing). All other institutions, including securitisation vehicles, are only available on a quarterly basis. The release of loan outstandings data for those lenders reporting on a quarterly basis will be lagged by one month - for example March outstandings for securitisation vehicles and other lenders n.e.c. will be released from the April publication onwards.
5 Until the statistics in this publication were derived from returns submitted to the Australian Prudential Regulation Authority (APRA) (see paragraph 9), the statistics of housing finance commitments covered all banks and permanent building societies. The largest of the remaining lenders of secured housing finance for owner occupation were included so that, together with banks and building societies, at least 95% of the Australian total of finance commitments were covered, and at least 90% of each state total was covered. While many smaller contributors to the Non-Banks series were excluded under these coverage criteria, at least 70% of finance commitments by wholesale contributors were covered.
6 When APRA commenced the collection, lending commitments by non-banks with total assets of $50 million or more were covered. All banks' lending commitments were covered.
7 From January 2014 a monthly reporting threshold was introduced for Non-Banks to provide 95 per cent asset coverage of the Non-Bank sector. Non-Banks with assets below the $200m asset threshold ceased reporting from January 2014 while other Non-Banks with assets above the threshold started reporting to APRA from January 2014. The lending commitments of those which started reporting in January 2014 were excluded from the January 2014 to January 2015 published statistics pending assessment of seasonal impacts of those Non-Banks' commitments on seasonally adjusted and trend series estimates. Issues of this publication from February 2015 onwards include finance commitments from January 2014 reported by Non-Banks above the reporting threshold. Non-Banks' Owner Occupied Housing Commitments are published in Tables 3 and 4. A trend break was added to the Non-Banks' series in January 2014 due to this change in coverage.
8 The statistics of housing loan outstandings cover all lenders included in the scope of paragraph 2 that have been identified as holding residential loan assets on their balance sheet as at the end of a particular reference month.
9 For banks, credit cooperatives, building societies and RFCs, the statistics in this publication are currently derived from returns submitted to the Australian Prudential Regulation Authority (APRA). The Financial Sector (Collection of Data) Act 2001 facilitates the collection of statistical data from the financial sector, with APRA established as the central point for collection of both prudential and statistical data. In October 2001, APRA implemented new reporting forms for building societies and credit cooperatives. New reporting forms were implemented for banks in March 2002, and for RFCs in March 2003. APRA commenced collecting loan commitments data from banks, credit cooperatives and building societies in July 2002, and from RFCs in March 2003.
10 Housing finance commitments for owner occupied housing from banks, building societies and credit cooperatives are derived from the ARF 392.0 Housing Finance form collected by APRA. Housing finance commitments for investor housing from these lenders are sourced from the ARF 394.0 Personal Finance form and the ARF 391.0 Commercial Finance form. Owner occupied housing finance commitments for RFCs are collected on the RRF 392.0 Housing Finance form. Investor housing commitments are collected on the RRF 394.0 Personal Finance form and the RRF 391.0 Commercial Finance form.
11 Statistics on loan outstandings in table 12 are sourced from banks on form ARF 320.0 Statement of Financial Position (Domestic Books) with lending by building societies and credit cooperatives derived from form ARF 323.0: Statement of Financial Position (Licensed ADI). While building societies and credit cooperatives with total assets greater than or equal to $200 million are required to report to APRA on a monthly basis, those institutions with total assets less than $200 million are only required to submit this return on a quarterly basis. An undercoverage adjustment is made in deriving table 12 in the two months between the last month in the quarter to derive estimates for the complete population on a monthly basis.
12 Electronic versions of the forms and instructions for ADIs are available on the APRA website at http://www.apra.gov.au/adi/ReportingFramework/Pages/reporting-forms-and-instructions-adis.aspx. For RFCs, these are available at: http://www.apra.gov.au/NonReg/Pages/Registered-Financial-Corporations.aspx.
13 All other institutions, including securitisation vehicles, are collected directly by the Australian Bureau of Statistics (ABS). Data on loan outstandings of households for housing purposes for these lender types are only available on a quarterly basis. The data for Other lenders n.e.c. is compiled from a range of other data sources collected by the Australian Bureau of Statistics (ABS).
14 Revisions to previously published statistics are included in the publication as they occur.
15 Changes in the classification of lenders (e.g. the conversion of a permanent building society to a bank) are reflected in the Lender series from the month of such change. Data for earlier periods for such lenders are not reclassified. Details of the establishment of new banks are recorded in the 'Series breaks' tabs of Statistical Table B2 on the Reserve Bank of Australia's website: RBA Statistical Tables.
16 A wholesale lender provides funds to borrowers through a retail intermediary which may then also be responsible for the ongoing relationship with the borrower.
17 The Wholesale Lenders n.e.c. series almost exclusively comprises securitisation vehicles (typically special purpose trusts), established to issue mortgage backed securities. It excludes commitments where a bank or permanent building society, acting as a wholesale provider of funds, is the lender on the loan contract. Those commitments are published as bank or permanent building society commitments.
18 From July 1995 to July 2000, mortgage managers reported housing finance commitments on behalf of wholesale lenders. The introduction of wholesale lenders as the reporting unit does not change the scope of the collection, but has increased its coverage. This, along with the reclassification of some lending activity, increased the level of the Wholesale Lenders n.e.c. series for owner occupied housing by $249m in July 2000.
19 Wholesale lenders contribute to the Non-Banks series for owner occupied housing, which is seasonally adjusted in table 3. A trend break was added to the Non-Banks series, shifting the trend up by 1,579 commitments and $178m in July 2000. Revisions related to the introduction of wholesale lenders also resulted in a downward shift in the Banks' trend for owner occupied housing of 1,256 commitments and $167m. Consequential breaks in the finance purpose trend series for owner occupied housing at July 2000 were:
20 Because of difficulties experienced by Wholesale Lenders n.e.c. in accurately identifying first home buyers in their commitments, these data are not used in estimating first home buyer commitments (table 9). Instead, from July 2000, the percentage of first home buyer commitments made by all banks and permanent building societies is applied to total Wholesale Lenders n.e.c. commitments to calculate their contribution to the First Home Buyers series. As a result, first home buyer commitments were revised upwards by 0.8 percentage points in July 2000.
21 An article on the introduction of the Wholesale Lenders n.e.c. series (including implications for the First Home Buyers series) featured in the October 2000 issue of this publication. A copy of the article is available from the ABS web site.
FIRST HOME BUYERS
22 First home buyers are persons entering the home ownership market for the first time. In 2014, it was established that some lenders were reporting only loans extended to first home buyers who had also received a First Home Owner Grant; instead of all first home buyers. The methodology used to adjust the estimates to account for the under-reporting was published in Information Paper: Changes to the method of estimating loan commitments to first home buyers, 2015 (cat. no. 5609.0.55.003) released on the ABS website on 4 February 2015.
23 The ABS and APRA worked successfully with lenders to ensure that all loans to first home buyers are reported, regardless of whether or not they received a First Home Owner Grant. As a result, from August 2016, the number of first home buyers no longer require adjustment as most lenders are reporting correctly. In the process of working with lenders, corrected historical data has been reported by some lenders and this improved data has been used to re-estimate the first home buyer statistics from July 2016 to October 2012. Information relating to these revisions and methods of estimating loans to first home buyers, can be found in the Information Paper: Changes to ABS First Home Buyer Statistics, Australia, 2016 (cat. no. 5609.0.55.004) released on the ABS website on 4 October 2016.
24 Seasonal adjustment is a means of removing the estimated effects of normal seasonal variation and ‘trading day effects’. A ‘trading day effect’ reflects the varying amounts of activity on different days of the week and the different number of days of the week in any month (i.e. the number of Sundays, Mondays, etc.). This effect may be partly caused by the reporting practices of the lenders. Adjustment is also made for Easter which may affect the March and April estimates differently. Trading day effects are removed from the original estimates prior to the seasonal adjustment process. Seasonal adjustment does not remove the effect of irregular or non-seasonal influences (e.g. a change in interest rates) from the series.
25 Over the period from early 1990 to April 1995, four of the major banks changed from reporting for the four or five weeks ending on the last Wednesday of each month to reporting on a calendar month basis. The published seasonally adjusted data take account of this change in pattern.
26 Rapid change in the financial sector, and particularly developments in the provision of housing finance, may cause changes in the seasonal and trading day patterns of the housing finance data. Examples include changes in the classification of financial institutions (particularly the reclassification of non-bank financial institutions to banks) and the increased use of mortgage securitisation.
27 Estimation of seasonal adjustment and trading day factors that reflect the full effect of recent developments is not possible until a sufficient number of years of data have been collected. When changes are occurring in the seasonal patterns, larger revisions to the seasonally adjusted series can be expected at the time of the biennial (once every two years) seasonal reanalysis. Accordingly, the trend estimate data provide a more reliable indicator of underlying movement in housing finance commitments. See paragraphs 30 and 31 for further information on trend estimates.
28 State component series have been seasonally adjusted independently of the Australian series. The sum of the state components in seasonally adjusted and trend series are therefore unlikely to equal the corresponding Australian totals. State component series are also affected by the changes mentioned in paragraphs 22 to 25.
29 The housing finance series uses a concurrent seasonal adjustment methodology to derive the seasonal adjustment factors. This means that original estimates available at the current reference month are used to estimate seasonal factors for the current and previous months. As a result of this methodology, the seasonally adjusted and trend estimates for earlier periods can be revised each month. However, in most instances, the only noticeable revisions will be to the previous month and the same month a year ago.
30 Autoregressive integrated moving average (ARIMA) modelling can improve the revision properties of the seasonally adjusted and trend estimates. ARIMA modelling relies on the characteristics of the series being analysed to project future period data. The projected values are temporary, intermediate values, that are only used internally to improve the estimation of the seasonal factors. The projected data do not affect the original estimates and are discarded at the end of the seasonal adjustment process. The lending finance collections use an individual ARIMA model for the majority of the series in this publication. The concurrent seasonal adjustment approach re-estimates seasonal factors each month with the receipt of each new observation. The parameters used for seasonal adjustment are routinely reviewed every 12 to 24 months to ensure the quality of the seasonal factors. The last reanalysis occurred in February 2017 for the January 2017 issue. For more information on ARIMA modelling see Feature article: Use of ARIMA modelling to reduce revisions in the October 2004 issue of Australian Economic Indicators (cat. no. 1350.0).
31 The best seasonally adjusted estimates are achieved only some years after corresponding original estimates have been released. However, this does not satisfy the demand for timely seasonally adjusted estimates. The ABS advises users that while every effort is made to achieve the highest possible quality of seasonally adjusted estimates, given the available original estimates and preset publication deadlines, revisions to these seasonally adjusted estimates are inevitable and generally indicate improvements to those estimates. The use of the concurrent seasonal adjustment approach means that revisions, and therefore quality improvements, are identified earlier than under the previously used forward factor method. Under the concurrent approach, revisions are made up to one year earlier than under the forward factor approach.
32 Smoothing seasonally adjusted series reduces the impact of the irregular component of the seasonally adjusted series and creates trend estimates. These trend estimates are derived by applying a 13-term Henderson-weighted moving average to all but the last six months of the respective seasonally adjusted series. Trend series are created for the last six months by applying surrogates of the Henderson moving average to the seasonally adjusted series. For further information, refer to Information Paper: A Guide to Interpreting Time Series-Monitoring Trends: An Overview (cat. no. 1349.0).
33 While the smoothing technique described in paragraph 30 enables trend estimates to be produced for the latest few months, it does result in revisions to the trend estimates as new data become available. Generally, revisions become smaller over time and, after three months, usually have a negligible impact on the series. Changes in the original data and re-estimation of seasonal factors may also lead to revisions to the trend.
EFFECTS OF ROUNDING
34 Where figures have been rounded, discrepancies may occur between sums of the component items and totals. Published changes in dollar value and percentage terms are calculated using unrounded data and may differ slightly from, but are more accurate than, changes calculated from the rounded data presented in this publication.
ABS DATA AVAILABLE ON REQUEST
35 Estimates for months prior to those shown in this publication and more detailed series are available in spreadsheet format from the ABS website - see the listing on pages 3 and 4. For more information, contact the ABS National Information and Referral Service on 1300 135 070.
36 Other ABS publications which may be of interest are outlined below. All publications released from 1998 onwards are available on the ABS website: https://www.abs.gov.au:
37 Quarterly data prior to March 2002 for housing loan outstandings by type of lending institution are available as a priced special data report related to the Australian National Accounts: Financial Accounts (cat. no. 5232.0). Inquiries regarding this special data report should be made to the contact on the front cover of this publication.
38 The Reserve Bank of Australia produces the quarterly Reserve Bank of Australia Bulletin as well as data on its website. Statistical Tables D1 and D2 contain data on lending and credit aggregates (including the housing credit aggregate). Table D5 Bank Lending Classified by Sector contains statistics on lending to persons for the purpose of housing, also classified by owner occupiers and investors, with statistics available from January 1990. Table B19 Securitisation Vehicles contains loans outstandings information for mortgages held, which includes both residential and non-residential mortgages.
39 APRA publishes residential lending by ADIs in Monthly Banking Statistics and Quarterly Authorised Deposit-taking Institution Performance.
40 Current publications and other products released by the ABS are available from the Statistics View. The ABS also issues a daily Release Advice on the ABS website https://www.abs.gov.au which details products to be released in the week ahead.
These documents will be presented in a new window.
Follow us on...Like us on Facebook Follow us on Twitter Follow us on Instagram