RDP 2021-10: The Rise in Household Liquidity Read me file

This ‘read me’ file contains information on the files used to generate the results presented in RDP 2021-10.

Data information

ABS

Some analysis presented in the paper comes from the Survey of Income and Housing (SIH) and the Household Expenditure Survey (HES). As confidential unit record data are used, the .dta files utilising these data are not publically available.

HILDA

Selected analysis presented in the paper comes from the Household, Income and Labour Dynamics in Australia (HILDA) Survey. The HILDA Survey provides individual-longitudinal level data. Data from the HILDA Survey is available to researchers living in Australia or overseas. Data may only be accessed through the National Centre for Longitudinal Data Dataverse. For more information on data access, please see <https://melbourneinstitute.unimelb.edu.au/hilda/for-data-users>.

‘Data’ folder

This folder contains the publically available data used by the .do files below.

Figure data

The file ‘rdp-2021-10-graph-data.xlsx’ contains the publically available data used to directly plot the figures. Data for Figure 3 uses APRA data that are confidential and not available for release.

Code

Our analysis was done using the statistical programming language Stata. We used version 16.0. The package ‘carryforward’ needs to be installed.

To produce the necessary output, .do files should be run in the order as listed below.

‘Replication – SIH’ – Survey of Income and Housing data

  1. Combine different waves of the SIH data

    Do file: Set up – Combine Household-level Data.do

    Functions: Combines key variables from raw SIH household-level data from 2003/04 to 2017/18 (except for 2007/08) and HES household-level data in 2003/04

    Output: 2004_2018.dta, 2004_2018withHES.dta

    Do file: Set up – Combine Loan-level Data.do

    Functions: Combines key variables from raw SIH loan-level data from 2007/08 to 2017/18 and HES loan-level data in 2003/04

    Output: loan_HES.dta

  2. Create variables

    Do file: Create Variables – Household Super.do

    Functions: Calculates household superannuation balances from SIH person-level 2017/18. The output is used for cross-sectional analysis

    Input: SIH17BP.dta

    Output: SIH17BP_household_super.dta

    Do file: Create Variables – Redraw Estimates.do

    Functions: Estimates redraw account balances with loan-level data

    Input: F5.dta, loan.dta, loan_HES.dta

    Output: redraw.dta, redraw_HES.dta

  3. Output for RDP figures

    Do file: Output – Cross-sectional graphs.do

    Functions: Creates data for Figure 4: Household Liquid Assets by Age of Household Head and Figure 5: Household Liquid Assets by Housing Tenure

    Input: SIH17BH.dta, SIH17BL.dta,SIH17BP_household_super.dta,redraw.dta

    Do file: Output – Time-series graphs.do

    Functions: Creates data for Figure 7: Household Liquid Assets; Figure 8: Household Liquid Assets by Housing Tenure; Figure 9: Hand-to-mouth Households; Figure 10: Hand-to-mouth Households by Housing Tenure; Figure 11: Household Liquid Assets and Figure 14: Household Liquid Assets

    Input: 2004_2018.dta, redraw.dta

    Do file: Output – Time-series graphs (with HES).do

    Functions: Creates data for Figure 12: Mortgage Debt and Household Liquidity and Figure 13: Mortgage Repayment

    Input: redraw_HES.dta, 2004_2018withHES.dta

  4. Output for RDP regressions (Table 3)

    Do file: Table 3 Regression – Set up.do

    Functions: Clean and creates variables for regression analysis

    Input: lendingrates.dta

    Do file: Table 3 Regression – Analysis.do

    Functions: Produces regression results for Table 3: The Effect of Interest Rates on Household-level Liquidity

‘Replication – HILDA’ – HILDA data

  1. Set up HILDA data

    Do file: HILDA – Merge – Individuals.do

    Functions: This file obtains unbalanced and balanced merged data files that contain the HILDA variables of interest of all persons in the HILDA surveys

    Do file: HILDA – Merge – Households.do

    Functions: This file loads the unbalanced HILDA individual file created in ‘HILDA – Merge – Individuals.do’ and returns an unbalanced and balanced household file

    Do file: Set up – Cash rate.do

    Functions: This program converts RBA statistical table F5 from excel format to dta format for later use

  2. Create variables

    Do file: HILDA – Create Variables.do

    Functions: This file uses the unbalanced HILDA household file created in ‘HILDA – Merge – Households.do’ and create variables used for Output – 1 and Output – 2

    Do file: HILDA – Create Variables – Income Uncertainty.do

    Functions: Creates ‘uncertainty.dta’ used for Output – 3

  3. Output for RDP

    Do file: HILDA – Output – 1.do

    Functions: Produces Table C2: Household-level Liquidity Buffers and Figure 14: Household Liquid Assets

    Do file: HILDA – Output – 2 .do

    Functions: Produces Table C1: Aggregate Household Liquidity Buffers, Statistical Area Level 4

    Do file: HILDA – Output – 3.do

    Functions: Produces Table 1: The Precautionary Saving Channel

    Do file: HILDA – Output -4.do

    Functions: Produces Table 2: The Housing Equity Withdrawal and Refinancing Channel

‘Replication – Other’ (input files found in ‘Data’)

Cross-country data

Do file: Cross-Country Evidence.do

Functions: Produces Figure B1: Response of Liquidity Buffer to a 1 Percentage Point Interest Rate Cut; Figure 2: Household Liquid Assets, OECD countries and Table B1: Change in Liquidity Buffer

Input: jorda.dta, Cross-Country Data.xlsx

Other Australian data

Do file: MP Shocks and Mortgage Payments.do

Functions: Produces Figure 18: Response of Mortgage Payments to Monetary Policy Shock

Input: Lending indicator.xlsx

Do file: MP Shocks and Liquidity Buffers – Financial Accounts.do

Functions: Produces Figure 17: Cumulative Response of Liquidity Buffer to Monetary Policy Shock

Input: 5206020_Household_Income.xls, 5232035.xls, rr_shocks.xlsx

10 November 2021

  • Supplementary information

Back to abstract