Our paper investigates the impact of government spending shocks on relative sector size and contrasts the effects across countries. Using a panel of sixteen OECD countries over the period 1970-2007, our VAR evidence shows that a rise in government consumption i) increases the share of non tradables in labor and real GDP while lowering the share of tradables, and ii) causes a significant increase in non traded wages relative to traded wages. While the first finding reveals that the non traded sector is more intensive in the government spending shock and experiences a labor inflow that increases its relative size, the second finding suggests the presence of labor mobility costs preventing wage equalization across sectors. These labor mobility costs appear to play a key role in determining changes in relative sector size across time and space. Whilst the responses of intersectoral labor reallocation and sectoral shares are found empirically to decline over time, the share of non tradables increases more in countries where the degree of labor mobility across sectors is higher. To account for our evidence, we develop an open economy version of the neoclassical model with tradables and non tradables. Our quantitative analysis shows that the open economy model is successful in replicating the responses of sectoral output shares to a fiscal shock, as long as we allow for a difficulty in reallocating labor across sectors along with adjustment costs to capital accumulation. Finally, calibrating the model to country-specific data, we are able to generate a cross-country relationship between the degree of labor mobility and the responses of sectoral output shares which is similar to that in the data. We are grateful to Agustin Benetrix for sharing RATS codes and to Luisito Bertinelli who provided invaluable help at different stages of the paper. We thank Benjamin Born, Falko Juessen, Gernot Müller for providing us with the dataset they constructed containing forecasts for government spending and Laura González Cabanillas for the dataset covering time series for the budget balance-GDP ratio forecast. We are grateful to Francisco Alvarez-Cuadrado, Christophe Hurlin, Stefan Schubert, Stephen Turnovsky for very helpful comments. We have also benefited from suggestions by participants at seminars and conferences. Obviously, any remaining shortcomings are our own. ,where 0 < ϕ J < 1 is the weight of the investment traded input and φ J corresponds to the elasticity of substitution between traded good and non traded good inputs.Since we focus on the long-run equilibrium, the tilde is suppressed for the purposes of clarity. the traded good and the non traded good according to a constant-returns-to-scale function which is assumed to take a CES form:,where ι is the weight of the investment traded input (0 < ι < 1) and φ J corresponds to the intratemporal elasticity of substitution in investment between traded and non traded inputs. The index J T is defined as a CES aggregator of home-produced traded inputs, J H , and foreign produced t...