انتخاب پرتفوی، ترجیحات رفتاری و جانبداری اصلی صاحبان سهام

We provide a plausible explanation of aggregate portfolio behavior, in a framework where economic agents have behavioral (narrow framing) preferences. The representative agent derives utility not only from consumption (standard models) but also from risky financial wealth fluctuations. Moreover, the investor frames the stock market risk narrowly and has loss averse preferences. We numerically solve, for the foreign equity share, a simple model of international portfolio choice, providing a possible explanation for the equity home bias puzzle. Only economic agents able to process correctly information deriving from stock markets exploit the diversification opportunities provided by international financial markets.

Recently, many contributions have emphasized that standard CRRA preferences have different problems in explaining some stock market puzzles1 and at the same time, behavioral theories have increasingly gained credit as an alternative explanation to these puzzles. In particular, some recent papers2 propose that people are loss averse over changes in the value of their stock market holdings. The basic idea is that, even if stock market risk is just one of many risks that determine their overall wealth risk, people still get utility directly from stock market fluctuations (narrow framing) and are more sensitive to losses than to gains (loss aversion). Christelis, Jappelli and Padula (2006) argue that individuals’ cognitive abilities may strongly affect investors’ financial choices, pointing out that cognitive ability is closely related to the ability to process information. In fact, evidence from psychology shows that poor cognitive skills are associated with low ability of processing information (Spaniol & Bayen, 2005): cognitive skills act as an additional constraint that optimizing individuals face when making their financial decisions. In order to illustrate one of the possible applications of behavioral preferences, we focus on the so-called equity home bias puzzle. Standard portfolio theory (mean/variance and consumption-based asset pricing models) states that it would be optimal for investors to hold a large fraction of their equity portfolio invested in foreign stocks3; but available empirical evidence is at odds with this theoretical prediction, showing that the most important components of household equity portfolios are domestic stocks: most countries hold a small share of foreign stocks in their equity portfolios. In particular, French and Poterba (1991) and Tesar and Werner (1995) estimated the percentage of aggregate stock market wealth invested in domestic equities in the beginning of the 1990s to have been well above 90% for U.S. and Japan,4 and around 80% for U.K. and Germany.5 During the 1990s the foreign equity participation by US investors has increased: Tesar and Werner (1998) show that in 1996 only around 10% of total U.S. equity holdings was invested abroad. But this level, if compared with what theoretical models predict, is too low. In the asset pricing/macroeconomics literature, many and different explanations have been provided about this puzzle. Lewis (1999) offers an extensive survey of potential explanations, ranging from the possibility for domestic stocks to better hedge home risks than foreign stocks, the presence of non-tradable consumption goods, diversification costs exceeding the gains, the effects of uncertainty about the economic environment and the role of measurement errors in the data. But Lewis concludes that “overall, equity home bias in portfolio levels remains a puzzle”. In other words, economists agree about the fact that, at the moment, no explanation is conclusive and fully satisfactory. In this paper, exploiting a behavioral finance based-approach,6 we argue that people with poor capabilities of processing information do not diversify their financial investments. And it is reasonable to suppose that, among individuals with poor capabilities of processing information, we find in particular people with a low level of education, while those with a higher level of education (an undergraduate degree or more) have higher capabilities. With this rationale in mind, we numerically solve a simple dynamic model of international portfolio choice, providing a possible explanation for the equity home bias puzzle. Barberis et al. (2006) solve a similar model finding numerically the parameter values for which an agent with a recursive utility function that allows for narrow framing would not participate in a stock market offering a high mean return and low correlation with other risks. But they do not solve explicitly the model neither for the equity asset share (in a context of stock market participation puzzle) nor for the foreign equity share (in a context of equity home bias puzzle). The paper is organized as follows. In Section 2 we discuss the basics of some behavioral finance models, finding their origins in the applied psychology literature. We also propose an original justification for the use of narrow framing preferences; moreover, we briefly see the Barberis–Huang (BH) model, which introduces the basic analytical formulations used in the paper. In Section 3, adopting the preferences introduced in the BH approach, we build and solve a simple international portfolio choice model in order to investigate and explain the equity home bias puzzle. The most important outcomes and implications of the model are presented and discussed in Section 4. Finally, Section 5 concludes the paper.

In this paper we have provided a contribution based on behavioral preferences, for explaining what we observe in the data about household equity portfolios. Standard portfolio theory states that for economic agents holding a foreign equity share higher than that actually held would be optimal. What happens is that convenient diversification opportunities are declined. Why? We find a satisfactory answer for this question by numerically solving a simple model of international portfolio choice, adopting the so-called “narrow framing” preferences. Basically, we stress that the mechanism in action is based on the individual's limited capabilities of processing information: foreign asset is perceived as less attractive than it would be if the investor had the optimal information skills and hence would be able to evaluate the two risky assets jointly. What follows is a low foreign equity share. The model matches available empirical evidence and is at odds with standard portfolio theory.22 We can also see the distinctive feature of this approach: in a descriptive context, it explains the actual (sub-optimal) choices of investors and not how they should behave in order to reach the optimum.