This paper examines the effect of implementation lags on government spending in a model with a cost channel. The main findings are summarized as follows. First, compared to an economy without implementation lags, the effect of implementation lags on the government spending multiplier depends on the degree of the cost channel. Second, the government spending multiplier becomes larger in an economy where long implementation lags and a strong cost channel are present. Third, the cost channel specification also changes the effect of implementation lags on the fiscal multiplier. Fourth, once we allow for the zero lower bound (ZLB) presence on the nominal interest rate, introducing a cost channel leads to a larger fiscal multiplier than in the case without a cost channel. Finally, introducing interest rate smoothing based on the the Taylor rule significantly affects the interaction effect of implementation lags and the cost channel on the fiscal multiplier in the model with and without the ZLB constraint.