This thesis introduces a novel dataset on bilateral de-jure exchange rate regimes. The new dataset accounts for the fact that officially pegging to one currency is uninformative about the exchange rate regime prevailing vis-á-vis other currencies, and it allows characterizing bilateral exchange rate regimes based on countries’ ex-ante announcements rather than ex-post observations. We use this data to estimate the effect of expected exchange rate volatility on foreign direct investment. Further, this thesis uses a new dataset on bilateral de-facto exchange rate regimes for the period 1973-2016 to study the effect of seven types of regimes on business cycle synchronization. Using the Extreme Bound Analysis methodology, we find that the exchange rate regime is a robust determinant of business cycle synchronization. Lastly, the thesis establishes a new effective exchange rate regimes classification. We show that the new effective approach is advantageous in order to correctly measure the effect of exchange rate regimes on inflation, because fixing an exchange rate regime to one currency does not completely anchor domestic prices in a multilateral world with large capital flows.