New Evidence And Perspectives On Household Investment Mistakes: I empirically test predictions of normative finance models using a unique set of disaggregated data on household inv ...
Senior Lecturer in Banking and Finance at Lincoln International Business School (UK)
New Evidence And Perspectives On Household Investment Mistakes: I empirically test predictions of normative finance models using a unique set of disaggregated data on household investment portfolios. Most findings of this study are consistent with stylized facts, but I uncover new mistakes and puzzles. On average, the households in the sample (i) perform well versus equity benchmarks but poorly against broader yardsticks; (ii) lack timing skill; (iii) tilt greatly towards regional stocks; (iv) eschew bond exposure; and (v) lowered their risk aversion before the financial crisis, and raised it subsequently. Many households exhibit profound re-balancing inertia, a behavior apparently not driven by lack of investment choice. Those that do re-balance display a small positive elasticity of risky share to wealth, consistent with DRRA. A minority of investors allocates with a 1/N spread, and this is associated with holding few funds. I record an unequivocal disposition effect, in the absence of tax distortions. Some householders fail to apply a simple and costless income tax reduction strategy.
Could Financial Driving Licenses Help Households Avoid Investment Mistakes?: Using a unique data set, I examine the relationship between investment outcomes of households and their financial acumen as assessed by an embryonic “financial driving license”. License-qualified households avoid novice mistakes but are still prone to more subtle biases. In the full sample, the most impactful mistakes observed are under-diversification, home bias, poor market timing, and disposition bias. The license system only weakly distinguishes composite aptitude, that is, vulnerability to serious losses; misclassification of vulnerable households is high. I find scant evidence to suggest that finance-related educational attainment or work experience predict investment skill. Rather, skill is most strongly associated with portfolio size (a wealth proxy) and age, neither of which the license under study employs. Certain categories of householders appear to purposely distort their portfolios, though not always profitably; CFA-holders outperform other groups.
|Thesis Committee :||
Supervisor: Ekkehart Boehmer, EDHEC Business School
External reviewer: Laurent Calvet, HEC Paris
Other committee member: René Garcia, EDHEC Business School