City Research Online

Abstract

This thesis identifies and quantifies the impact of cognitive errors in certain aspects of investor decision-making. One error is that investors are unaware that the Internal Rate of Return (IRR) is a biased indicator of expected terminal wealth for any dynamic strategy where the amount invested is systematically related to the returns made to date. This error leads investors to use Value Averaging (VA). This thesis demonstrates that this is an inefficient strategy, since alternative strategies can generate identical outturns with lower initial capital. Investors also wrongly assume that the lower average purchase cost which is achieved by Dollar Cost Averaging (DCA) results in higher expected returns. DCA is a similarly inefficient strategy.

Investors also adopt strategies such as Volatility Pumping, which appears to benefit from high asset volatility and large rebalancing trades. This thesis demonstrates that any increase in the expected geometric mean associated with rebalancing is likely to be due to reduced volatility drag, and that simpler strategies involving lower transactions costs are likely to be more profitable. Academic papers in highly-ranked journals similarly misinterpret the reduction in volatility drag achieved by rebalanced portfolios, mistakenly claiming that it results from the rebalancing trades “buying low and selling high”.

The previously unidentified bias in the IRR has also affected an increasing number of academic studies, leading to misleadingly low estimates of the equity risk premium and exaggerated estimates of the losses resulting from bad investment timing. This thesis also derives a method for decomposing the differential between the GM return and the IRR into (i) the effects of this retrospective bias, and (ii) genuine effects of investor timing. Using this method I find that the low IRR on US equities is almost entirely due to this bias, and so should not lead us to revise down our estimates of the equity risk premium. This method has wider applications in fields where IRRs are used (e.g. mutual fund performance and project evaluation).

In identifying these errors this thesis makes a contribution: (i) to the academic literature by correcting previous misleading results and improving research methods; (ii) to investment practitioners by identifying avoidable errors in investor decision-making. It also makes a contribution to the field of behavioural finance by altering the range of investor behaviour which should be seen as resulting from cognitive error rather than the pursuit of different objectives.

Publication Type:	Thesis (Doctoral)
Subjects:	H Social Sciences > HG Finance
Departments:	Bayes Business School > Finance Doctoral Theses Bayes Business School > Bayes Business School Doctoral Theses

Export

Downloads