Research Discussion Paper – RDP 2019-06 The Effect of Mortgage Debt on Consumer Spending: Evidence from Household-level Data

This ‘read me’ file contains details of the code and data included in this archive that were used to generate the results reported in RDP 2019-06. Publically available plotting data for figures appearing in the RDP can be found in the spreadsheet ‘rdp-2019-06-graph-data.xls’.

If you make use of any of these files you should clearly attribute the authors in any derivative work.

Data

The following data sources were used:

• Household-level data:
  • Obtained from Department of Social Services; Melbourne Institute of Applied Economic and Social Research (2018), ‘The Household, Income and Labour Dynamics in Australia (HILDA) Survey, RESTRICTED RELEASE 17 (Waves 1-17)’, doi:10.26193/OFRKRH, ADA Dataverse, V2.
• Aggregate household debt (including and excluding offset account balances):
  • Internal RBA data – not available for release.
• Aggregate household final consumption expenditure and income:
  • Obtained from the Australian Bureau of Statistics (ABS) Australian National Accounts: National Income, Expenditure and Product, Australia, Cat No 5206.0, March 2019
• Population:
  • Obtained from the Australian Bureau of Statistics (ABS) Australian Demographic Statistics, Cat No 3101.0, December 2018
• Consumer price index:
  • Obtained from the Australian Bureau of Statistics (ABS) Consumer Price Index, Australia, Cat No 6104.0, March 2019
• Postcode-level housing prices:
  • Obtained from CoreLogic – not available for release
• Local unemployment rate:
  • Obtained from the Australian Bureau of Statistics (ABS) Labour Force, Australia, Detailed – Electronic Delivery, Cat No 6291.0.55.001, May 2019

Figure data

Data for all figures except the top panel of Figure 1 are publically available. Aggregate household debt data including and excluding offset account balances are not publically available.

The underlying data files as used by the code referenced in all do files below are not included in this archive due to the terms of our access; as such, the code for all files will not run.

Code

The results reported in this RDP were generated using Stata 13.0.

Included in this archive are the following programs:

• 1. HILDA merge code – individuals.do
• 2. HILDA merge code – households.do
• 3. Set up variables.do
• 4. Main models and Appendix E.do
• 5. Main models – interactions.do
• 6. Spending normalisation.do
• 7. Aggregate effect.do
• 8. Appendix C.do
• 9. Appendix F.do
• 10. Graph data.do

The first three programs should be run in the order indicated by the numerical prefix of the file. The first program extracts the relevant data from the Department of Social Services Household, Income and Labour Dynamics in Australia (HILDA) Dataverse (not provided in this documentation) to create a longitudinal person-level file. The second program identifies households to create a longitudinal household-level file. The third program creates all the necessary variables for the analysis.

The remaining programs provide the output for the Tables and Figures in the paper. ‘4. Main models and Appendix E.do’ uses the longitudinal file to estimate the benchmark results in Tables 1, 2 and 6 and Appendix E (Tables E1–E3). ‘5. Main models – interactions.do’ provides the results for Tables 3, 4 and 7, while the results for Table 5 can be found using ‘6. Spending normalisation.do’. ‘7. Aggregate effect.do’ estimates the results for Table 8.

The results in Appendix C (Table C1) and Appendix F (Table F1) can be found using ‘8. Appendix C.do’ and ‘9. Appendix F.do’, respectively. The estimates in Figures 1–6 that are based on household-level data can be found using ‘10. Graph data.do’.

Back to abstract