Search

We estimate a novel measure of global financial uncertainty (GFU) with a dynamic factor framework that jointly models global, regional, and country-specific factors. We quantify the impact of GFU shocks on global output with a VAR analysis that achieves set-identification via a combination of narrative, sign, ratio, and correlation restrictions. We find that the world output loss that materialized during the great recession would have been 13% lower in absence of GFU shocks. We also unveil the existence of a global finance uncertainty multiplier: the more global financial conditions deteriorate after GFU shocks, the larger the world output contraction is.

We investigate the determinants of sovereign CDS spreads on a sample of Eastern European data. A dynamic hierarchical factor model is used to aggregate information in indicators of economic fundamentals. CDS spreads are regressed on forecasts of factors. We find that domestic fundamentals explain more of CDS spread variance than global factors, largely due to their ability to explain differences in sovereign risk across countries. The effects on CDS spreads are found to be time-varying. In terms of economic significance, the factor of institutional-political strength stands out. We apply the model to study CDS spreads of Poland, Russia and Turkey.

O presente trabalho explora se a curva de juros brasileira, especialmente seu fator de longo prazo, via modelo NS, responde a variáveis internacionais, mais especificamente a choques em variáveis econômicas, financeiras ou, mesmo, no fator de longo prazo da economia dos Estados Unidos. Inflação global, produção industrial global e fatores latentes internacionais fazem parte dos dados escolhidos. Para tal, é extraída uma curva de juros global de quatro países (Estados Unidos, Canadá, Reino Unido e Brasil) através de um Dynamic Hierarchical Factor Model. O modelo possui diversas vantagens, como considerar a propriedade temporal dos dados e permitir tanto a estimação da curva global em um único passo, além da obtenção coeficientes que possam nos retornar de que forma variáveis respondem aos choques. Os resultados indicam que a curva de juros global está em um nível inferior ao nível da curva de juros brasileira, que essa responde aos choques em variáveis externas, principalmente aos dados ligados a taxa de juros supranacionais, e que o efeito em direção contrária não é simétrico. The present paper explores whether the Brazilian yield curve, especially its long-term factor, via the NS model, responds to international variables, more specifically, shocks in economic variables, financial variables or even the long-term factor of the United States economy . Global inflation, global industrial output and international latent factors are part of the data chosen. To do so, a global yield curve is extracted from four countries (United States, Canada, United Kingdom and Brazil) through a Dynamic Hierarchical Factor Model. The model has several advantages, such as considering the temporal property of the data and allowing both estimation of the global curve in a single step, as well as obtaining coefficients that can inform us in what way variables respond to shocks. The results indicate that the global yield curve is at a level below the level of the Brazilian yield curve, which responds to shocks in external variables, mainly to the data related to supranational interest rates, and that the effect in the opposite direction is not symmetrical.

This doctoral dissertation addresses the problem of how to measure a latent process such as globalisation which manifests itself in a large number of statistical indicators. A new globalisation measure is offered. It is based on the evaluation of proportion of future economic growth explained by foreign indicators relative to domestic ones. The construction of this measure is based on the assumption that there is a strong relationship between economic openness or globalisation and economic growth. A method to evaluate the offered globalisation measure is developed so that the design of the model is suited for the structure of the data. Targeted indicators are structured into domestic and foreign blocks and included into a dynamic hierarchical factor model. The model is evaluated employing Markov chain Monte Carlo with Gibbs sampling. The resulting factors are included into economic growth forecast equation and time-varying parameters loading of domestic and foreign factors are evaluated by applying Kalman filter on a dynamic linear model. The application for Lithuanian economy revealed a positive trend in foreign impact on the economy and it depicts the globalisation effect.

This doctoral dissertation addresses the problem of how to measure a latent process such as globalisation which manifests itself in a large number of statistical indicators. A new globalisation measure is offered. It is based on the evaluation of proportion of future economic growth explained by foreign indicators relative to domestic ones. The construction of this measure is based on the assumption that there is a strong relationship between economic openness or globalisation and economic growth. A method to evaluate the offered globalisation measure is developed so that the design of the model is suited for the structure of the data. Targeted indicators are structured into domestic and foreign blocks and included into a dynamic hierarchical factor model. The model is evaluated employing Markov chain Monte Carlo with Gibbs sampling. The resulting factors are included into economic growth forecast equation and time-varying parameters loading of domestic and foreign factors are evaluated by applying Kalman filter on a dynamic linear model. The application for Lithuanian economy revealed a positive trend in foreign impact on the economy and it depicts the globalisation effect.

In this paper we reconsider the degree of international comovement of inflation rates. We use a dynamic hierarchical factor model that is able to decompose Consumer Price Index (CPI) inflation in a panel of countries into (i) a factor common to all inflation series and all countries, (ii) a factor specific to a given sub-section of the CPI, (iii) a country group-factor and (iv) a country-specific component. With its pyramidal structure, the model allows for the possibility that the global factor affects the country-group factor and other subordinated factors but not vice versa. Using quarterly data for industrialized and emerging economies from 1996 to 2011 we find that about two thirds of overall inflation volatility are due to country-specific determinants. For CPI inflation net of food and energy, the global factor and the CPI basketspecific factor account for less than 20% of inflation variation. We argue that local inflation rather than global inflation (Ciccarelli and Mojon (2010)) is a better description of the evidence. Only energy price inflation in industrial economies is dominated by common factors.

The present paper examines the degree of comovement of gross capital inflows, which is a highly sensitive issue for policy makers. We estimate a dynamic hierarchical factor model that is able to decompose inflows in a sample of 47 economies into (i) a global factor common to all types of flows and all recipient countries, (ii) a factor specific to a given type of capital inflows, (iii) a regional factor and (iv) a country-specific component. We find that the latter explains by far the largest fraction of fluctuations in capital inflows followed by regional factors, which are particularly important for emerging markets' FDI and portfolio inflows as well as bank lending to emerging Europe. The global factor, however, explains only a small share of overall variation. The exposure to global drivers of capital flows, i.e. the global factor and the factor specific to each type of capital inflows, is particularly pronounced for countries with a more developed financial system. A fixed exchange rate regime does not shield countries from the ebb and flow of global capital flow cycles.