To what extent can the policy of hoarding international reserves in indebted and dollarized countries be efficient ?

Summary The first chapter of this thesis studies the efficiency of IR accumulation and sterilization in dollarized and indebted countries, by measuring the sterilization coefficient, and the reversal coefficient. This chapter explores the link between the sources of reserves and external debt. By applying the 2SLS regression model, we identify the explanatory variables that allow us to estimate the mentioned coefficients. The results obtained indicate that despite the theoretical and correct application of the sterilization policy, economic constraints contribute to reduce the expected efficiency of monetary policies. The second chapter considers the probabilities of indebted countries falling into financial crises despite the fact that they accumulate international reserves acting as a moderating shock and/or self-insurance. We use the Financial Stress Indicator (FSI), proposed by Balakrishnan et al (2009) and the IMF, which covers the various aspects of financial crises. We apply the Markov Switching model with varying probability. We obtain the result that debts increase the probability that a country suffers from a financial crisis, however, IRs do not necessarily provide "peace" in the economy, except in a few cases. However, the negative effects of debt outweigh the positive effects of IR, especially in the relatively more dollarized countries. The third chapter first measures the degree of the trilemma indices: exchange rate stability, monetary independence and capital account openness, while taking into account the accumulation of IR to GDP or External Debt (ED) ratios in the short run. The evolution of the trilemma indices shows that countries that adopt a "de facto" flexible exchange rate take advantage of the benefits of ERs to adopt an administered exchange rate regime, which consists of simultaneously achieving the three objectives of the trilemma without giving up any of them. Interpretations may change if IRs are taken as a function of debts, i.e., the use of IRs/debt should be considered in such studies. Second, we find that for countries that adopt a de facto fixed exchange rate regime, IRs (different ratios) play no role in the evolution of the Mundell triangle and do not play a role in the monetary policy decisions of the monetary authorities. Finally, this chapter discusses the normative aspect of the trilemma, linking policy choices to macroeconomic outcomes such as output growth volatility. We note that the results are different across countries, and depend on different IR measurement ratios. We conclude that the impact of IRs on output growth volatility may change depending on the level of EDs and on the exchange rate regime adopted.