On Endogenous Market Incompleteness, Cycles, and Growth

Abstract

Esta tesis doctoral consiste en tres ensayos independientes sobre Macroeconomía y Crecimiento Económico.<br/><br/>Ensayo 1. "Technological Transfers, Limited Commitment and Growth"<br/>Las evidencias muestran que hay sustanciales flujos de capital de países ricos hacia países pobres, aunque no son tan abundantes como sugieren las diferencias en tasas de rendimiento. Esos flujos son pro-cíclicos: abundantes en tiempos de bonanza y escasos en los malos tiempos. Los modelos de crecimiento convencional afrontan ciertas dificultades para explicar estos hechos. En este estudio, proponemos un modelo dinámico de flujos de capital hacia países en desarrollo que es cualitativamente consistente con regularidades empíricas. El modelo está basado en tres premisas principales: i) los contratos de préstamo internacional son imperfectamente ejecutables; ii) el acceso a los mercados financieros internacionales está relacionado con transferencias tecnológicas a un país en desarrollo del resto del mundo; iii) algunos de los beneficios tecnológicos asociados al acceso a la financiación externa son perecederos. Nuestras conclusiones sugieren que las transferencias tecnológicas pueden desempeñar el papel de mecanismo que hace cumplir las obligaciones contractuales.<br/><br/>Essay 2. "A Note on Computing Partial Derivatives of the Value Function by Simulation"<br/>Los problemas que contienen restricciones de compatibilidad de incentivos han recibido una amplia atención en la literatura debido a los recientes avances en optimización dinámica. Frecuentemente las condiciones de optimalidad para este tipo de problemas contienen derivadas parciales de la función de valor (value function) con respecto a algunas variables de estado endógenas. En este trabajo proponemos un algoritmo para calcular estas derivadas parciales por simulación. <br/><br/>Essay 3. "Institutions and Growth: Some Evidence from Estimation Methods Partially Robust to Weak Instruments"<br/>Este estudio se centra en la estrategia empírica propuesto por Hall y Jones (1999) para estimar el efecto de lo que ellos denominan "infraestructura social" sobre la productividad de los países. Nosotros intentamos responder la crítica de Acemoglu et al (2001) a esta metodología por basarse en instrumentos geográficos. Para ello consideramos el problema de identificación débil en los modelos de variables instrumentales de Hall y Jones (1999). La evidencia obtenida de los estimadores parcialmente robustos como k-class y jackknife está interpretada en base a los estudios de Monte Carlo. Podemos concluir que el uso de algunos estimadores k-class permite utilizar variables lingüísticas para instrumentar la calidad institucional a pesar de su baja correlación con el regresor endógeno en cuestión.

This doctoral thesis consists of three self-contained essays in Macroeconomics and Economic Growth.<br/><br/>Essay 1. "Technological Transfers, Limited Commitment and Growth"<br/>Evidence shows that there are substantial rich-to-poor international capital flows although not as abundant as differences in rates of return would suggest. These flows are procylcical: abundant in good times and scarce in bad times. They have been reported to promote growth and stability in some countries, but merely to augment instability in the others. Conventional growth models face certain difficulties in accounting for this pattern. In this paper, we propose a dynamic model of capital flows to developing countries which is qualitatively consistent with these empirical regularities. The model is based on three main premises: i) international lending contracts are imperfectly enforceable; ii) access to the international financial markets results in technological transfers to a developing country from the rest of the world; iii) some of the productivity gains associated with the access to external financing are perishable. We solve for transitional dynamics of the model economy with endogenously incomplete markets and compare the results with the solutions obtained from the perfect risk-sharing and autarkic environments. In addition, we examine the implications of alternative assumptions about the severity of the repudiation punishment for growth, welfare and borrowing patterns. Our findings suggest that technological transfers may play a role of an important enforcement mechanism. In our framework, existence of substantial rich-to-poor capital flows is not inconsistent with the presence of default risk.<br/><br/>Essay 2. "A Note on Computing Partial Derivatives of the Value Function by Simulation"<br/>The problems involving incentive compatibility constraints in the form of dynamic participation constrains have received wide attention in the literature due to the recent advances in dynamic optimization techniques. Often the optimality conditions for this class of problems involve partial derivatives of the value function with respect to some of the endogenous state variables. In this paper we suggest an algorithm for computing these partial derivatives by simulation. The attractive features of the algorithm include its rather wide scope of applicability and simplicity of implementation. Furthermore, the suggested method does not suffer from the curse of dimensionality and therefore it is particularly convenient for the models involving many state variables.<br/><br/>Essay 3. "Institutions and Growth: Some Evidence from Estimation Methods Partially Robust to Weak Instruments"<br/>This paper focuses on the empirical approach proposed by Hall and Jones (1999) to estimate the effect of what they call "social infrastructure" on productivity across countries. We attempt to address the criticism of Acemoglu et al (2001) directed towards this methodology for relying on the geographical instruments. To do so we consider the issue of weak identification in the linear instrumental variables model of Hall and Jones (1999). The evidence obtained from the partially robust estimators like the k-class and jackknife estimators is interpreted on the basis of the Monte Carlo studies. Our findings suggest that using some of the k-class estimators allows exclusive reliance on the linguistic variables to instrument for institutional quality despite their low correlation with the endogenous regressor in question.