We propose that stock-market participation is inf luenced by social interaction. In our model, any given "social" investor finds the market more attractive when more of his peers participate. We test this theory using data from the Health and Retirement Study, and find that social households-those who interact with their neighbors, or attend church-are substantially more likely to invest in the market than non-social households, controlling for wealth, race, education, and risk tolerance. Moreover, consistent with a peer-effects story, the impact of sociability is stronger in states where stock-market participation rates are higher.The Journal of Finance in the stock market. On the one hand, if one takes the perspective of a fullinformation, frictionless model with optimizing households-in which those who sit out do so simply because they find the market's risk-return profile unattractive-standard arguments suggest that there is nothing to be gained by having the government invest in the market on their behalf. On the other hand, if households do not participate because of a lack of information about market opportunities, or because other frictional costs deter them from doing so, the case for these proposals can, at a minimum, begin to make logical sense. 2 A few basic facts about the determinants of household participation in the stock market are already well known. 3 First, participation is strongly increasing in wealth. This can be understood by thinking of participation as involving fixed costs; wealthier households have more to invest, and so the fixed cost is less of a deterrent to them. Vissing-Jorgensen (2000) builds a model in which fixed costs of participation are incurred on a per-period basis, and estimates that such costs have to be on the order of $200 per year to explain observed participation rates. Of course, such large numbers beg the questions of what these black-box fixed costs actually represent, and whether one should think of them as being similar across different types of households.Stock-market participation has also been found to be increasing in household education. One interpretation is that education reduces the fixed costs of participating, by making it easier for would-be investors to understand the market's risk-reward trade-offs, to deal with the mechanics of setting up an account, and executing trades, etc. 4 Finally, there is also a pronounced link between race and participation, with white, non-Hispanic households having much higher participation rates, controlling for wealth and education.In this paper, we add to this line of work by considering the possibility that stock-market participation is inf luenced by social interaction. A priori, this would seem to be a promising hypothesis, given the growing body of empirical research that speaks to the importance of peer-group effects in a variety of other contexts. 5 Notably, some of this work finds evidence of peer effects in financial settings that are suggestively close to the one we have in mindfor example, Madrian and Shea...