The switch from a fixed exchange rate regime to a flexible exchange rate regime seldom goes smoothly. A major reason why devaluations are so disruptive is that countries are reluctant to abandon their fixed exchange rate regimes. This “reluctance to devalue” phenomenon is one of the puzzles in international finance. This dissertation makes towards understanding this “reluctance to devalue”. First, I investigate the factors that may influence the probability of a switch from a fixed to a flexible exchange rate regime using survival models. I find that pegs have non-monotonic duration dependence. Moreover, I find that GDP growth strongly influences the probability of abandoning a peg. Second, I propose that the “reluctance to devalue” could stem from uncertainty about the control over inflation after devaluation which raises the threshold of economic pain that could convince policy makers to devalue. I develop this argument in a rules-vs-discretion theoretical framework. Empirical analysis based on survey data from Bulgaria supports this hypothesis. Given that abandoning a fixed exchange rate regime is one of the three options that are available to countries on a peg, I investigate whether a periphery country's decision to abandon its peg is impacted by a potential move to a currency union. I find that the perception of “insurance” justified by expected-bailouts in a currency union increase the support for joining a currency union. The strength of this “safety net” perception is strong despite expected negative impact of the currency union on the country’s macroeconomic indicators.