We incorporate reference-dependent worker behavior into a search-matching model of the labor market, in which firms have all the bargaining power and productivity follows a log-linear AR(1) process. Motivated by Akerlof (1982) and Bewley (1999), we assume that existing workers’ output falls stochastically from its normal level when their wage falls below a "reference point", which (following K˝oszegi and Rabin (2006)) is equal to their lagged-expected wage. We formulate the model game-theoretically and show that it has a unique subgame perfect equilibrium that exhibits the following properties: existing workers experience downward wage rigidity, as well as destruction of output following negative shocks due to layoffs or loss of morale; newly hired workers earn relatively flexible wages, but not as much as in the benchmark without reference dependence; market tightness is more volatile than under this benchmark. We relate these findings to the debate over the “Shimer puzzle” (Shimer (2005)).