Five Facts about Beliefs and Portfolios

Abstract

We study a newly designed survey administered to a large panel of wealthy retail investors. The survey elicits beliefs that are important for macroeconomics and finance, and matches respondents with administrative data on their portfolio composition, their trading activity, and their login behavior. We establish five facts inthese data. (i) Beliefs are reflected in portfolio allocations. The sensitivity of portfolios to beliefs is small on average, but varies significantly with investor wealth, attention, trading frequency, and confidence. (ii) Belief changes do not predict when investors trade, but conditional on trading, they affect both the direction and the magnitude of trades. (iii) Beliefs are mostly characterized by large and persistent individual heterogeneity. Demographic characteristics explain only asmall part of why some individuals are optimistic and some are pessimistic. (iv) Expected cash flow growth and expected returns are positively related, both within and across investors. (v) Expected returns and the subjective probability of rare disasters are negatively related, both within and across investors. These five facts provide useful guidance for the design of macro-finance models. (JEL D83, E23, G11, G12, G41, G51)

Belief Disagreement and Portfolio Choice

“…Specifically, by identifying an event that changes beliefs and controlling for the main channels through which the event could have changed portfolios other than through beliefs, we show that heterogeneous beliefs drive both reported beliefs and differential portfolio allocations. While survey beliefs are qualitative, the combination of a large swing in beliefs and a small change in behavior is in line with the attenuation puzzle, that is, the low correlation between portfolio holdings and reported beliefs across investors, documented by Ameriks et al (2020), and the finding that changes in reported beliefs do not predict the likelihood of trading but do predict the direction of trading conditional on a trade, documented by Giglio et al (2021).…”

“…(Color figure can be viewed at wileyonlinelibrary.com) While these results present the data in a transparent manner, they do not show that changes in beliefs were the driving force for these reallocations rather than differences in preferences or differential changes in hedging needs due to differential changes in the stochastic process of labor income or the local economy. Much existing research relates reported beliefs to economic behaviors (see Manski (2018)), and to portfolio choice in particular (e.g., Vissing-Jørgensen (2003), Ameriks et al (2020), Giglio et al (2021)). In general, however, many factors change both behavior and beliefs, and thus, may contaminate any analysis of the effect of beliefs on behavior.…”

A New Test of Risk Factor Relevance

“…There are survey papers asking questions about consumption growth (Di Maggio, Kermani, and Majlesi (2020)) and investors' views on popular finance models (Choi and Robertson (2020), Giglio et al. (2021), Liu et al. (2022)).…”

“…Prior work has looked at how people frame financial decisions-for example, over individual positions (Odean (1998)), across positions (Frydman, Hartzmark, and Solomon (2018)), and over portfolios (Hartzmark (2015)). There are survey papers asking questions about consumption growth (Di Maggio, Kermani, and Majlesi (2020)) and investors' views on popular finance models (Choi and Robertson (2020), Giglio et al (2021), Liu et al (2022)). We use a surveybased framework that shows that, contrary to textbook logic, investors' framing of their portfolio decision does not involve consumption-growth correlations as an input.…”

A Fine-Grained Annotated Corpus for Target-Based Opinion Analysis of Economic and Financial Narratives

“…An investor buys or sells an asset based on his view about its expected future price. A recent study, for example, finds some evidence that investors' beliefs can be reflected in respondents' multi-asset allocation strategy (Giglio et al, 2021).…”