the composition of government spending and the real exchange rate 11
4. Conclusion
The main goal of this paper has been to show that the composition of government spend-
ing matters for the long-run dynamics of the real exchange rate. In our theoretical model,
we have shown that government consumption typically leads to real appreciation but
that the sensitivity of the real exchange rate to government investment depends on how
an increase in the public capital stock differentially affects sectoral productivity levels in
the traded and nontraded sectors. Our empirical estimates confirmed that government
consumption indeed generates real appreciation for the full sample, while showing that
government investment has no significant long-run impact on the real exchange rate for
the set of EMU member countries.
Next, we checked whether the fiscal variables operate through the relative price of
nontradables in the manner embedded in the theoretical analysis. We found that an in-
crease in government consumption raises the relative price of nontradables for the EMU
sample and sub-samples, while government investment is associated with a decline in
the relative price of nontradables for most country groups. In turn, we found that the
productivity gains from an increase in government investment are concentrated in the
nontraded sector, which is consistent with the relative price evidence and the mecha-
nism specified in our theoretical model.