The thesis investigates the extent to which Pareto inefficiency may occur as a result of individuals working more than that which is socially optimal. Many of the most important contributions to the literature are discussed and presented, starting with an analysis of externalities resulting from individuals' concern regarding their social status, which is illustrated by a simple model of conspicuous consumption. Several existing models of imperfect information in which agents work too much are subsequently presented. Examples with either moral hazard or adverse selection, or both, are analysed, including models in which agents have career concerns. After a detailed discussion of many of the previous contributions to the literature, two alternative models are developed. We first consider a simple model of adverse selection in teams in which agents differ solely by their productivity. The market equilibrium involves equal work input from all agents, which implies that a fraction of the population works too little, whereas the remaining fraction works too much. It is shown that the efficient allocation can be sustained under an appropriate policy intervention. Next a shirking-model is presented in which employees are induced to exert effort due to the fear of losing their job. Workers' contribution to output is a function of both their effort level and their amount of work hours. Corresponding to similar findings in the existing literature, our model involves an inefficiently low level of employment. However, under the assumption of decreasing marginal productivity, we also find that the sub-optimal level of employment may cause firms to require their employees to work longer hours than the social planner's first-best choice. We also state a sufficient condition under which the market determination of individual work hours is always sub-optimally high, even when the information costs are taken into account. Having found that individual overemployment is very much a possible outcome of the market determination of individual work hours, we also discuss the efficiency merits of restricting hours worked. Even when employees report working more than their desired level at the prevailing wage rate, a policy restricting work hours may push the economy off the locus of Pareto efficient equilibria without increasing worker welfare. The potential ineffectiveness of restricting work hours as a means to boost employment is briefly illustrated using the previously developed shirking-model.