| Summary: | This thesis is divided into eight main chapters and makes contributions to the area of financial crises and international finance. The first chapter provides a general introduction to the thesis and highlights the main contributions of our work. The second chapter is a literature review which provides a well-defined structure to organise our thoughts about the literature on micro-systemic risks and Central Bank policy. The chapter initially reviews the literature for single-bank crises. It then proceeds on to provide a succinct account of multiple-bank crises and of financial crises that result from the interaction between banks and financial markets. The main value-added of this chapter is that it helps us identify those areas in the literature in which research work is missing. This provides legitimate foundation for building new models to address these issues. The subsequent chapters of this thesis have emerged to bridge these missing gaps identified in our literature review. Chapters 3, 4 and 5 deal with banking panic transmission in two-bank scenario. With common investments affected by the same macroeconomic fundamental, a crisis that spreads from one bank to another has contagious and correlated elements. The purpose of these chapters is to provide a robust theoretical account that can enable us distinguish between these two elements in probability terms. We embed a two-bank model within a dynamic Bayesian setting and use the global games approach to derive the existence of trigger equilibrium in each bank. Chapter 3 provides an overview of our banking environment. Chapter 4 makes a contribution to derivation of the equilibrium concept and shows the equivalence between Perfect Bayesian Equilibrium (PBE) of our game and trigger equilibrium. Chapter 5 encapsulates all results. We show the existence of contagion as one of `excess correlation' between banks. This allows us to depart from existing theoretical papers which explain contagion as interdependence. Furthermore, we show that whether contagion or correlation occurs is a function of the relative importance depositors attach to private vs public signals. The chapter ends by identifying some puzzles (zero-link, clustering and avoidance) which our paradigm can explain and throws light on ways (which are not captured by single-bank models) Central Banks should implement prudential policy measures. Chapters 6, 7 and 8 deal with financial crises in open economies. An important limit of existing bank run models is that they are developed without taking into account the level of economic development of the economy in which they occur. We are interested in studying how financial crises occur in an Emerging Market Economy (EME) and, most importantly, how the nature of their occurrence differs when the exogenous macroeconomic constraints of an EME are duly accounted for. Chapter 6 introduces the main banking environment we study in the subsequent two chapters. Important among the assumptions are that all depositors in the banking system are foreign investors (the economy is fully liberalised) and that banks have balance sheets characterised by liability dollarisation. We use the mechanism design approach to show the existence of a pecking order in allocation of resources and liquidity. In particular, we show that a banking allocation is weakly Pareto- inferior to that of a Planner who observes all structures of the economy but its stochastic fundamentals and who achieves second-best allocation. Once this banking allocation is derived, Chapter 7 studies the nature of the transmission mechanism from a banking crisis to a currency crisis. We show that under certain parametric restrictions, Lender-of-Last Resort (LOLR) policies may be a conduit that generates a currency crisis. In a multi-bank setting, LOLR may even be sub-optimal since it may induce devaluation-based bank runs at other banks. Chapter 8 studies the reverse causation from a currency crisis to a banking crisis. Both chapters 7 and 8 offer useful guidance to policy. The success of a policy measure depends on its ability to restore the Planner's second-best. A number of policy options (e.g state-contingent controls) are studied and suggestions for design of exchange rate regimes, based on ability to ward off the twin crises, are offered as well.
|
|---|
