非洲经济研究中心-南非消费、房价、抵押贷款和非抵押贷款债务的新模式（英）

New models for South African consumption, house prices, and mortgage and non-mortgage debt Insights for financial stability and monetary transmission JANINE ARON CSAE, Department of Economics and INET, University of Oxford, UK and JOHN MUELLBAUER INET and Nuffield College, University of Oxford, UK October 2025 Abstract: Aggregate consumption typically exceeds 60 per cent of GDP and should be pivotal in central bank policy models. Most use semi-structural macro-models, yet consumption is usually inadequately specified. We use a systems approach to estimate new equations for South African consumption, house prices, mortgage and non-mortgage debt, and income forecasting. A credit-augmented consumption function approach introduces a greater role for uncertainty and a key role for credit conditions, and varies the spendability of different wealth components. This provides new insights into the multiple monetary transmission mechanisms, from policy interest rates and credit conditions to aggregate demand, including via non-homogeneous household balance sheet items on consumption. Credit conditions for mortgages and for other debt move quite differently from each other, with implications for consumer spending. Non-mortgage debt covers a larger fraction of total household debt than in advanced market economies, affecting household financial vulnerability. Housing market participants tend to extrapolate recent house price changes when forming expectations of capital gains, so positive shocks to housing demand can feed back onto house prices and consumption and extend boom conditions. House prices and debt can overshoot relative to their fundamentals, affecting financial stability. These findings should benefit future policy modelling in South Africa. Key words: aggregate consumption, house prices, credit conditions, household debt, housing collateral, monetary transmission, equation system JEL classification: C32, E21, E51, E58 Acknowledgements: We thank Wian Boonzaaier, Elriette Botes, Rashad Cassim, Shaun de Jager, Barend De Beer, Riaan Ehlers, Karen Kuhn, Danie Meyer, Caswell Monyela, Mpho Moloto, Lesego Morope, Khumbudzo Muneri, Susana Paulse, Pieter Pienaar, Bart Stemmet, and Rowan Walter at the South African Reserve Bank (SARB) for their advice, particularly on data. We also thank Greg Farrell, John Loos (First National Bank), Illana Melzer (71point4), Siphamandla Mkhwanazi (First National Bank), Hendrik Nel, Shannon Bold (Bureau for Economic Research), Johan van den Heever, and Ben Smit. We are grateful to Nick Burger and Melissa Meisel for extensive discussions and for access to data by Lightstone; and to Hermine Bester (Nedbank) for kindly providing data from her Master’s thesis on repeat-sales indices for house prices. Finally, we are grateful to Philippe Burger for his insightful comments, which have improved the paper. This study has been prepared within the UNU-WIDER project Southern Africa—Towards Inclusive Economic Development