Essays on monetary policy frameworks

Abstract

This thesis presents three essays on monetary policy frameworks (MPFs) within emerging and developing countries. In the first essay I examine the choice of MPF of 87 countries over three decades (1985–2017). I split the factors that determine MPF choice into three thematic groups: optimum currency area (OCA), financial strength, and political/institutional strength. By using a multinomial logit model, I confirm the importance of a range of factors including financial development, trade networks, strength of democracy, and size of the economy. Furthermore, I use two different cross-validation methods to find that the model is good at predicting emerging and developing countries’ choice of MPF. In the second essay, I analyse how MPF choice impacts foreign direct investment (FDI) inflows for 83 countries. estimate the average treatment effects on the treated (ATT) using four different approaches: (1) inverse probability weighting (IPW) with propensity scores estimated via logistic regression, (2) treatment-weighted estimation using logistic propensity scores, (3) treatment-weighted propensity scores estimated via generalised boosted modelling (GBM) using the twang package, and (4) double debiased machine learning (DDML) with stacking. The preferred model—DDML with stacking—estimates that D-framework countries receive, on average, 2.84 percentage points less FDI (as a share of GDP) than ER-framework countries. This difference is not only statistically significant, but also economically meaningful, exceeding the average annual increase in FDI inflows for developing countries during key benchmark years. In the third essay, I see how MPF affects exchange rate pass-through (ERPT) for four emerging countries that are not major oil exporters: Brazil, Chile, Israel, and South Africa. This essay analyses how a change in MPF, from pre-inflation targeting (pre-IT) to inflation targeting (IT) affects ERPT. I employ a time-varying parameter VAR model with stochastic volatility (TVP-VAR-SV). This allows for a more flexible representation of gradually evolving macroeconomic relationships, including the dynamics of ERPT. I generate 3D impulse response plots separately for each regime to assess whether changes in the MPF are associated with shifts in ERPT. The 3D impulse response functions (IRFs) indicate that, across all four countries, the adoption of IT is associated with a decline in ERPT, although the extent and pattern of this decline vary by country. Moreover, some of the observed shifts in the IRFs correspond closely with the timing of MPF changes as classified in the classification, lending additional credibility to the results.