Education on Participation by SES and Predictors of Exit Decisions of Stock Owners from the National Longitudinal Survey of Youth 1979

Abstract

The stock market participation puzzle has been widely studied in the household finance literature, but there are interesting details that are still left unexplored. With the National Longitudinal Survey of Youth 1979 (NLSY79) data, I study active participation at the cross section and dynamically. I show that one more year of schooling predicts active stock market participation well only for those of mid socioeconomic status (mid SES), even controlling for perfect foresight of future income as well as a host of demographic characteristics. One more year of schooling does not affect participation for those in low SES and high SES, which reflects how respectively, one group tends not to participate while the other tends to invest in stocks. I also show that income shock, which is theorized to operate as a liquidity constraint, is a good predictor of the decision to exit the stock market, or the decision to liquidate all stock holdings in non-retirement accounts, in the short run, while the probability of exiting in the long run is hard to predict accurately with income shock. Other proxies of liquidity constraint, changes wealth and family related changes, do not predict either exit decisions well.