Financial Analyst Judgement

Abstract

Financial analysts are frequently required to exercise professional judgement on a wide variety of financial considerations including security returns, corporate growth rates, bankruptcy risks and merger/takeover prospects. Typically these financial decisions require the careful accumulation and skilled interpretation of a multitude of descriptive and predictive financial statistics in. addition to qualitative and other non-qualitative information. The combined effects of these diverse and intercorrelated courses of financial and non-financial information is then' determined and subsequently incorporated into some decision model or action plan. Although a considerable volume of theoretical and empirical research has been undertaken on the capital markets at the aggregate or market level, the decision behaviour of the individual analysts who operate in these markets have received only scant attention in the business literature. Within this context, the purpose of this study was to provide some preliminary evidence on the properties and characteristics of financial analyst judgement at the individual analyst level The major research instrument consisted of a. laboratory experiment in which thirty eight experienced financial analysts assessed the expected risk and return characteristics of an individual security in a well diversified portfolio. Subject responses were conditional on the value of nine financial and capital market variables identified via a factor analysis with varimax rotation plus one industry instrumental variable. Data were extracted from actual publicly listed New Zealand companies and background information was provided to enhance task realism. The results of this study indicated that whilst financial analysts were moderately to highly reliable in formulating their judgement evaluations (i.e. high intra-rater reliability) there was considerable heterogeneity in financial analyst judgement evaluations between subjects (i.e. low inter-rater consensus). Furthermore an analysis of individual analyst multiple regression models indicated this inter-rater disagreement was not random, but resulted from systematic differences in subject cue utilization strategies. Additional analyses revealed (1) financial analysts possessed only moderate insight into their cue utilization strategies (2) a composite judge provided an effective means of reducing the inconsistency and instability inherent in individual financial analyst judgement and (3) a psychologically determined measure of subject risk propensity was largely ineffectual in explaining individual differences in financial analyst judgement evaluations. The results of the main experiment were cross validated via a separate but related take-home judgement task.