{"title":"Stock Market Participation and Portfolio Shares Over the Life-Cycle","authors":"Francisco Gomes, O. Smirnova","doi":"10.2139/ssrn.3808350","DOIUrl":null,"url":null,"abstract":"We estimate the life-cycle profile of stock market participation and risky portfolio share. We address the classical identification problem by running the estimations in first differences, which allows us to estimate the age profiles without making any assumptions on time or cohort effects. We find that stock market participation is a hump-shaped function of age, increasing early in life and decreasing after age 60. The conditional risky share also decreases late in life, but it is a flat function of age before that. We also investigate the economic mechanisms driving this behavior. Our results provide empirical support for the importance of participation costs in explaining stock market participation, and for models where investors have decreasing relative risk aversion and where human capital is a close substitute for bonds, although not completely uncorrelated with stock returns. Finally, background risks are also likely to play a role, particularly late in life. We conclude by presenting a structural life-cycle model that closely replicates our empirical results.","PeriodicalId":354235,"journal":{"name":"ERN: Behavioral Life Cycle Models (Topic)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"ERN: Behavioral Life Cycle Models (Topic)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.2139/ssrn.3808350","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 5
Abstract
We estimate the life-cycle profile of stock market participation and risky portfolio share. We address the classical identification problem by running the estimations in first differences, which allows us to estimate the age profiles without making any assumptions on time or cohort effects. We find that stock market participation is a hump-shaped function of age, increasing early in life and decreasing after age 60. The conditional risky share also decreases late in life, but it is a flat function of age before that. We also investigate the economic mechanisms driving this behavior. Our results provide empirical support for the importance of participation costs in explaining stock market participation, and for models where investors have decreasing relative risk aversion and where human capital is a close substitute for bonds, although not completely uncorrelated with stock returns. Finally, background risks are also likely to play a role, particularly late in life. We conclude by presenting a structural life-cycle model that closely replicates our empirical results.