Kuilen, G. van de (Tilburg University, Center for Economic Research)
We conduct an experiment to study the prevalence of the higher order risk attitudes of prudence and temperance, in a large demographically representative sample, as well as in a sample of undergraduate students. Participants make pairwise choices between lotteries of the form proposed by Eeckhoudt and Schlesinger (2006). The choices in these lotteries isolate prudent from imprudent, and temperate from intemperate, behavior. We relate individualsâ€™ risk aversion, prudence, and temperance levels to demographics and financial decisions. We observe that the majority of individualsâ€™ decisions are consistent with risk aversion, prudence, and temperance, in both the student and the demographically representative sample. An individualâ€™s level of prudence is predictive of his wealth, saving, and borrowing behavior outside of the experiment, while temperance predicts the riskiness of portfolio choices. Our findings suggest that the coefficient of relative prudence for a representative individual is approximately equal to two.
Recent contributions have shown that it is possible to account for the so-called consumption-real exchange anomaly in models with goods market frictions where international asset trade is limited to a riskless bond. In this paper, we consider a more realistic international asset market structure and show that as soon as we depart from the single bond economy, we can no longer account for the consumption-real exchange anomaly. Our central result holds for a simple asset market structure in which two nominal bonds are traded across countries. We explore the role of demand shocks such as news shocks in generating meaningful market incompleteness. We show that only under specific settings news shocks can improve the performance of the model in matching the portfolio positions and consumption-real exchange rate correlations that we observe in the data.
|Keywords:||Portfolio choice, incomplete financial markets, international risk sharing, consumption-real exchange rate anomaly|