Three essays on the impact of inflation targeting in developing and emerging countries

The objective of this dissertation is to estimate the effects of adopting inflation targeting (IT) on inflation and economic growth in developing countries. From an econometric perspective, the main shortcoming of most studies on IT effects is selection bias: the choice of implementing IT is intrinsically endogenous, in which case traditional regression techniques may yield biased estimates. To address this issue, each essay applies a novel methodology borrowed from the impact evaluation literature that has seldom been used in the monetary field.The first essay implements the synthetic control method to compare the macroeconomic trajectory of individual inflation targeters with that of a combination of comparison economies. Most of the results regarding inflation are unusable due to the sui generis nature of this series in developing countries. Evidence on the impact of IT on output is more insightful, albeit mixed. In Chile, Ghana, and Peru, real GDP is higher than that of their respective synthetic controls after the adoption of IT. However, the opposite is found in other countries, such as Mexico and Guatemala, and inconclusive results are found for various other countries.The second essay seeks to estimate IT effects by using a combination of entropy balancing, a weighting scheme designed to improve covariate balance between IT and non-IT countries, and fixed effects. This identification strategy helps to address potential endogeneity issues while keeping the benefits of the temporal nature of the dataset. While the evidence on the gains in terms of inflation reduction is mixed, we find that IT adoption tends to reduce output growth.The third essay aims to assess the impact of IT by using an endogenous treatment effect approach. This methodology enables us to jointly estimate the probability of implementing IT and the outcome equation, thereby addressing endogeneity issues. The analysis of the results confirms the existence of an output-inflation tradeoff under IT.Overall, the results of these estimates suggest caution in the application of IT policies. Especially, monetary authorities should recognize that there could potentially be output costs of any gains in reduced inflation and adjust the implementation of IT policies accordingly.