Statement on Monetary Policy – November 2019 Box C
Updates to the Financial Aggregates

Credit and money, otherwise known as the financial aggregates, are important indicators of households' and businesses' financial activity. The Reserve Bank compiles and publishes estimates of the financial aggregates in Australia. Since July 2019, these have been based on an improved data collection known as the Economic and Financial Statistics (EFS) collection.^[1]

The credit aggregates measure funds borrowed by Australian households and businesses from financial intermediaries. Total credit includes housing credit, business credit and personal credit (such as credit cards). The monetary aggregates measure banknotes and coins held by households and businesses, and ‘money-like’ instruments – such as deposits – that households and businesses hold with financial institutions.

Over the past few years, the Australian Prudential Regulation Authority (APRA), the Australian Bureau of Statistics and the Reserve Bank have worked to modernise the existing data collected from banks and other reporting financial institutions and used to compile the financial aggregates (and a number of other measures of financial institutions' activities and performance). The new EFS collection has more precisely defined data, including for loans, which are now better classified by their purpose, rather than by the type of collateral or nature of the borrower.

While the new EFS collection has resulted in more comprehensive and accurate estimates of the credit and money aggregates, it has not changed the Reserve Bank's assessment of the growth rates of credit or money. While there are noticeable changes to the levels of housing credit, personal credit and money aggregates, these are not material when measured as a share of GDP or disposable income. The level of business credit was largely unchanged.

Growth rates are similar under the new collection

Credit growth has not been affected in any noticeable way by the move to the new EFS collection in July 2019 (Graph C.1). We know this because between March and June 2019, banks and other institutions reported using the old and new data collections in parallel.

During this period, extensive data quality assessments and engagement with industry were undertaken to ensure that the new financial aggregates based on the EFS collection provide a good indication of developments in financial activity.

The level of housing credit is estimated to be a little lower

The new definitions and more comprehensive guidance in the EFS collection led to a shift in the estimated levels of the outstanding stocks of several aggregates between June and July 2019.^[2] Banks and other institutions reclassified around $25 billion of housing credit as personal credit, reflecting the definitions and guidance around loan purpose (Graph C.2). For example, in the previous collection, topping up a housing loan to buy a car may have been recorded as a housing loan. However, such loans are now recorded as personal loans if the predominant purpose of the loan is personal finance, such as buying a car.

The estimate of the level of housing credit is also lower by around $30 billion due to more accurate measurement of housing credit extended by non-authorised deposit-taking institutions (non-ADIs). Recent legislative changes led many non-ADIs to register with APRA and, therefore, report data on their lending activities. Previously, the financial aggregates included an estimate of lending by non-ADIs that did not report to APRA. This estimate was based on lending by a smaller sample of these lenders, scaled up by an estimate of their share in total lending. The more comprehensive and more accurate sample of non-ADI lenders in the new data suggests that the earlier estimate was too high. The estimates based on the EFS data continue to suggest that non-ADIs account for less than 5 per cent of housing lending.

Some owner-occupier loans were reclassified as investor loans

The composition of housing credit has also changed with the new EFS collection. Investor housing credit as a share of total housing credit increased by 5 percentage points (Graph C.3). This is because owner-occupier housing loans are now defined as loans for the principal place of residence only, such that a borrower can generally only have one owner-occupier loan. This is not necessarily how banks and other institutions had previously recorded housing loans, which may have resulted in some loans previously recorded as owner-occupier – for example, a loan for a holiday house – being reclassified as investor loans under the new EFS collection.

There has also been some reclassification within the monetary aggregates

The monetary aggregates can be broken down into currency (that is, banknotes and coins held by the private sector), transaction deposits (which can be easily used to make payments) and non-transaction deposits (such as term deposits). The EFS collection resulted in $120 billion in non-transaction deposits being reclassified as transaction deposits, as the collection more accurately defines deposits by their type (Graph C.4). For example, in the previous collection, reporting institutions had classified deposits in some online savings accounts as non-transaction deposits.

In addition, the definition of M1 has been expanded to fully capture the more liquid forms of money. Previously, M1 included only currency and chequing accounts at banks. M1 now includes currency and all transaction deposits at banks, credit unions and building societies (Graph C.5).^[3] This new definition has also been applied to the historical data for M1.

Endnotes

For more information see Bank J, K Durrani and E Hatzvi (2019), ‘Updates to the Financial Aggregates’, RBA Bulletin, March. [1]

Consistent with usual practice, the growth rates for July were adjusted for the effects of these level changes. The Reserve Bank publishes growth rates (which are break-adjusted) and the levels of the credit aggregates (which are not adjusted) in Statistical Tables D1 and D2 on the RBA website. Growth rates should not be calculated from data on the level of credit. [2]

See Bank J, K Durrani and E Hatzvi (2019), ‘Updates to the Financial Aggregates’, RBA Bulletin, March. The Reserve Bank publishes five different monetary aggregates: currency, the money base, M1, M3 and broad money. These measures differ in the types of instruments that are included. For example, M1 contains only the most liquid forms of money such as banknotes and current deposits (i.e. funds that can be easily used to make payments), whereas broad money also includes term deposits and short-term debt securities issued by financial institutions in Australia. [3]