This paper builds a bridge between the two existing approaches for wage and employment determination in a unionized market: the monopoly union model and the efficient bargaining model. Both fail to capture the dynamic aspects of wage bargaining. When the repeated nature of the wage bargaining process is considered, the equilibria are neither as inefficient as the monopoly union model predicts nor as fully efficient. Rather, the two models can be regarded as particular cases with certain discount rates. We apply our model to issues such as the endgame interpretation of the U. S. steel industry, wage concessions, and featherbedding.