Household investment mistakes are an important concern for researchers and policymakers alike. Portfolio underdiversification ranks among those mistakes that are potentially most costly. However, its roots and empirical importance are poorly understood. I estimate quantitatively meaningful diversification statistics and investigate their relationship with key variables. Nearly all households that score high on financial literacy or rely on professionals or private contacts for advice achieve reasonable investment outcomes. Compared to these groups, households with below-median financial literacy that trust their own decision-making capabilities lose an expected 50 bps on average. All group differences stem from the top of the loss distribution.ECONOMIC THEORY PREDICTS THAT households will hold their risky assets in the form of a well-diversified portfolio. The extent to which this prediction holds true has important implications for the regulation of consumer financial products, the design of retirement savings plans, and the distribution of consumer well-being in general. It is especially significant to know the patterns of underdiversification. If only those households that are particularly savvy in financial matters invested their retirement wealth in a few highly correlated stocks, most policymakers would conclude that this was a rational choice driven, for example, by superior information (van Nieuwerburgh and Veldkamp (2010)). If only the least financially literate households were doing the same thing, however, policy makers would likely conclude that these households were simply making poor decisions. In a similar vein, a regulator's strategy also depends on * Von Gaudecker is with Universität Bonn, Department of Economics. Several of the data manipulations and preliminary estimations were performed by Mark Prins as part of his MSc Thesis at VU University Amsterdam and in a subsequent research assistantship to the author. Special thanks to him for an excellent programming job and many fruitful discussions.

We analyse risk preferences using an experiment with real incentives in a representative sample of 1,422 Dutch respondents. Our econometric model incorporates four structural parameters that vary with observed and unobserved characteristics: Utility curvature, loss aversion, preferences towards the timing of uncertainty resolution, and the propensity to choose randomly rather than on the basis of preferences. We find that all four parameters contribute to explaining choice behaviour. The structural parameters are significantly associated with socio-economic variables, but it is essential to incorporate unobserved heterogeneity in each of them to match the rich variety of choice patterns in the data. JEL: C90, D81

Objectives Examine the effects of the COVID-19 pandemic on the mental health and loneliness in the general population. More specifically, the study focused on prevalence of anxiety and depression symptoms, the extent to which individuals with existing symptoms recovered or not, the prevalence of subtypes of loneliness, and the extent to which loneliness before and during this pandemic was associated with anxiety and depression symptoms. Methods Data was extracted from the longitudinal LISS panel, based on a probability sample of the Dutch population, with assessments on loneliness in October 2019 (T1) and June 2020 (T4), and anxiety and depression symptoms in November 2019 (T2), March 2020 (T3) and June 2020 (T4; Ntotal = 4,084). Loneliness was examined with the De Jong Gierveld Loneliness Scale and anxiety and depression symptoms with the Mental Health Inventory (MHI-5). Results Repeated measures multivariate logistic regression analyses (RMMLRA) showed a statistical significant lower prevalence of anxiety and depression symptoms after the outbreak (T4 = 15.3%) than before (T2 = 16.8%) and during the COVID-19 outbreak (T3 = 17.2%). According to the Reliable Change Index, the distribution of recovery categories (remission, improvement, unchanged and worsening symptoms) after the outbreak did not differ significantly from the distribution of these categories before the outbreak. RMMLRA revealed that the prevalence of emotional loneliness increased significantly after the outbreak (T1 = 18.4%, T4 = 24.8%). Among individuals who were not lonely before and after the outbreak the prevalence of symptoms decreased significantly (T2 = 7.0%, T4 = 4.4%) and, likewise, among those who were not lonely anymore after the outbreak (T2 = 21.5%, T4 = 14.5%). However, the prevalence of symptoms increased significantly among those who became lonely during the pandemic (T2 = 17.9%, T4 = 26.3%). Conclusions Findings suggest that this pandemic did not negatively affect the prevalence of anxiety and depression symptoms nor the normal recovery of symptoms among the general population during the first four months, but that emotional loneliness increased.

The Impact of Income Shocks on Health:Evidence from Cohort Data *We study the effect of permanent income innovations on health for a prime-aged population. Using information on more than half a million individuals sampled over a twenty-five year period in three different cross-sectional surveys we aggregate data by date-of-birth cohort to construct a 'synthetic cohort' dataset with details of income, expenditure, socio-demographic factors, health outcomes and selected risk factors. We then exploit structural and arguably exogenous changes in cohort incomes over the eighties and nineties to uncover causal effects of permanent income shocks on health. We find that such income innovations have little effects on a wide range of health measures, but do lead to increases in mortality and risky health behaviour.JEL Classification: I1, J3

No abstract