Precautionary motives and portfolio decisions
- 9 October 2002
- journal article
- research article
- Published by Wiley in Journal of Applied Econometrics
- Vol. 18 (1) , 61-77
Abstract
This paper studies the empirical relevance of precautionary and other motives for household portfolio behaviour using recent panel data from the Netherlands. Dutch households' portfolios exhibit low degrees of risk taking and diversification. It is possible that this is the outcome of a rational, precautionary response to unavoidable exposure to background risk (stemming from the labour market or health conditions, etc.). We consider as alternative explanations liquidity needs and habits. The endogenous variable is the fraction of clearly safe in total financial assets at the household level. Parametric and semi‐parametric censored regression models for pooled cross‐sections and random and fixed effects models for panel data show that both heteroscedasticity and unobserved heterogeneity are of major importance in the data. With subjective indicators of income uncertainty we find a limited role for precautionary motives. Copyright © 2002 John Wiley & Sons, Ltd.Keywords
This publication has 24 references indexed in Scilit:
- Extensive Margins and the Demand for Money at Low Interest RatesJournal of Political Economy, 2000
- Portfolio Choice and Asset Prices: The Importance of Entrepreneurial RiskThe Journal of Finance, 2000
- Estimation of a censored regression panel data model using conditional moment restrictions efficientlyJournal of Econometrics, 2000
- Income variability, uncertainty and housing tenure choiceRegional Science and Urban Economics, 1999
- Consumption and Portfolio Selection with Labor Income: A Continuous Time ApproachMathematical Finance, 1998
- Standard Risk AversionEconometrica, 1993
- Testing for Selectivity Bias in Panel Data ModelsInternational Economic Review, 1992
- Trimmed Lad and Least Squares Estimation of Truncated and Censored Regression Models with Fixed EffectsEconometrica, 1992
- Symmetrically Trimmed Least Squares Estimation for Tobit ModelsEconometrica, 1986
- Least absolute deviations estimation for the censored regression modelJournal of Econometrics, 1984