The author presents empirical evidence that he has obtained from an analysis of the response of different economic variables, including the real wage rate, to a technology shock. He replicates Galí's (1999) bivariate model and compares dynamic impulse responses and conditional correlations with evidence provided by the vector-error-correction model that was identified using the King, Plosser, Stock, and Watson (1991) procedure. To calculate confidence intervals, the author uses Kilian's (1998) bootstrap-after-bootstrap method. The empirical evidence suggests that it is not possible to reject a procyclical real wage in response to a technology shock. Therefore, real-business-cycle models cannot be rejected based on their conditional predictions of the labour-market dynamics in favour of other types of models.