Download this article in PDF format
The way investment risk is perceived by the actors on the capital markets strongly depends on the hypotheses and conventions that have been adopted as standards in the measurement of risk. The fact that certain standards have been used since long doesn’t necessarily mean that they are the most adequate and accurate. This paper points at a number of those standards that are liable to lead to a misperception of risk. Conventions are being reviewed that are profoundly inscribed in the theory and practice of investment management. It is not new to alert for the limits of financial modelling. Shortcomings in financial models and concepts are much discussed in the literature, and are regularly subject of debate. In most studies the limitations of a model are of secondary order though after the presentation of the model itself. Few research projects have been undertaken that assess the validity of the standard axioms in finance as the primary objective. Such study makes one realise to what extent the perception of investment risk is shaped by the conventions that have been established over time. It helps to explain why the investment industry seems weak to respond to novel circumstances and why it is prone to systemic failure. The goal of this paper is this, to help understand the weaknesses in the finance industry in terms of risk perception. Analyses underlying this paper have been carried out in a traditional investment setting. Common risk concepts are analysed that apply to common investments instruments. No derivatives, credit risk vehicles or other structured investment products are considered, for which the profit-and- loss profile is not directly intuitive. Well-established finance models are used as the reference for defining the investment risk, notably the Capital Asset Pricing Model (CAPM) introduced by Sharpe (1964) and the Arbitrage Pricing Theory (APT) introduced by Ross (1976). It is not evident to perceive risk; even less is it easy to perceive risk perception, as the act of perception is in the same time subjective, irresolute and unequivocal. Perception is a cognitive experience that starts by a form of measurement which necessarily simplifies the world such that it enables an analytical assessment which eventually leads to some form of judgment.
Marielle de JONG
Head of Fixed Income Quantitative Research at Amundi
The standard indices available for the bond investment markets, are composed of securities that are weighed by the size of the outstanding debt, and are for that reason weighing heavy on the ones most indebted. Those market indices are risky by construction and indeed appear to be. They are shown to be mean-variance inefficient in recent studies, which is disconcerting since that is in contradiction with the principle underlying the Capital Asset Pricing theory that the market as a whole is in price equilibrium. If market-weighted indices are inefficient, market weighting is inefficient! We contribute to a growing literature on this question, which mostly focuses on equities, by testing on bonds. In the construction of market indices we weigh by the fundamental value of the debt issuers rather than their debt size. We do this for sovereigns using Gross Domestic Product figures, and for corporates taking sales revenues. The tests, which we run in the Eurozone over the sovereign debt crisis, add to the evidence that market weighting may indeed be inefficient.
Marielle de JONG, Hongwen WU