We use a two-sector neoclassical open economy model with traded and nontraded goods and endogenous markups to investigate the effects of temporary fiscal shocks. One central finding is that theory can be reconciled with evidence once we allow for endogenous markups and assume that the traded sector is more capital-intensive than the nontraded sector. More precisely, although both ingredients are essential to produce the real exchange rate depreciation, only the second ingredient is necessary to account for the simultaneous decline in investment and the current account, in line with the evidence.