Evaluating the existence of structural change in the brazilian term structure of interest: evidence based on cointegration models with structural break

This paper investigates whether there is evidence of structural change in the Brazilian term structure of interest rates. Multivariate cointegration techniques are used to verify this evidence. Two econometrics models are estimated. The rst one is a Vector Autoregressive Model with Error Correction Mechanism (VECM) with smooth transition in the deterministic coe¢ cients (Ripatti and Saikkonen [25]). The second one is a VECM with abrupt structural change formulated by Hansen [13]. Two datasets were analysed. The rst one contains a nominal interest rate with maturity up to three years. The second data set focuses on maturity up to one year. The rst data set focuses on a sample period from 1995 to 2010 and the second from 1998 to 2010. The frequency is monthly. The estimated models suggest the existence of structural change in the Brazilian term structure. It was possible to document the existence of multiple regimes using both techniques for both databases. The risk premium for di¤erent spreads varied considerably during the earliest period of both samples and seemed to converge to stable and lower values at the end of the sample period. Long-term risk premiums seemed to converge to inter-national standards, although the Brazilian term structure is still subject to liquidity problems for longer maturities.

Isolated capital cities and misgovernance: theory and evidence

Motivated by a novel stylized fact { countries with isolated capital cities display worse quality of governance { we provide a framework of endogenous institutional choice based on the idea that elites are constrained by the threat of rebellion, and that this threat is rendered less e ective by distance from the seat of political power. In established democracies, the threat of insurgencies is not a binding constraint, and the model predicts no correlation between isolated capitals and misgovernance. In contrast, a correlation emerges in equilibrium in the case of autocracies. Causality runs both ways: broader power sharing (associated with better governance) means that any rents have to be shared more broadly, hence the elite has less of an incentive to protect its position by isolating the capital city; conversely, a more isolated capital city allows the elite to appropriate a larger share of output, so the costs of better governance for the elite, in terms of rents that would have to be shared, are larger. We show evidence that this pattern holds true robustly in the data. We also show that isolated capitals are associated with less power sharing, a larger income premium enjoyed by capital city inhabitants, and lower levels of military spending by ruling elites, as predicted by the theory.

Workplace diversity and incentive contracts : theory and evidence

This paper presents a simple theory of the provision of incentives in firms in which the principal optimally chooses both compensation contracts and the composition of the work force. Assuming that individuals display group loyalty, a less diverse (more homogeneous) work force will be more cooperative. Simple comparative statics provide some testable implications relating risk, diversity and incentive pay. I also analyze the case in which workers’ characteristics cannot be readily observed ex ante. The theory then predicts that firms are more likely to prevent workers from interacting with each other when workers are expected to have similar characteristics. This shows a surprising effect of diversity in the workplace: more diverse firms will promote more interactions between workers of different types, i.e. they will be less segregated. I test the main predictions of the model using a cross-sectional sample of corporate boards. I use the proportion of women on boards as a measure of diversity. There are three main empirical findings: (1) a significant negative correlation between firm risk and diversity, (2) a significant positive relationship between performance-based compensation and diversity and (3) a significant positive correlation between the number of board meetings (a measure of interactions among directors) and diversity. The evidence is broadly consistent with the implications of the theory.

The influence of lifestyle and gender on sickness absence in brazilian workers

Background: Despite an increasing body of knowledge concerning gender and lifestyle factors as determinants of sickness absence in well-developed countries, the relationship between these variables has not been elucidated in emerging economic power countries, where the burden of non-communicable diseases is particularly high. This study aimed to analyze the relationships among lifestyle-related factors and sick leave and to examine whether gender differences in sickness absence can be explained by differences in socio-demographic, work and lifestyle-related factors among Brazilian workers. Methods: In this longitudinal study with a one year follow-up among 2.150 employees of a Brazilian airline company, sick leave was the primary outcome of interest. Independent variables collected by interview at enrolment in the study were gender, age, educational level, type of work, stress, and lifestyle-related factors (body mass index, physical activity and smoking). In addition, the risk for coronary heart disease was determined based on measurement of blood pressure, total cholesterol and glucose levels. The total number of days on sick leave during 12 months follow-up was available from the company register. Logistic regression analysis was used to determine the influence of socio-demographic, type of work and lifestyle-related factors on sick leave. Results: Younger employees, those with lower educational level, those who worked as air crew members and those with higher levels of stress were more likely to have sick leave. Body mass index and level of physical activity were not associated with sick leave. After adjustment by socio-demographic variables, increased odds for 10 or more days of sick leave were found in smokers (OR = 1.51, CI = 1.05-2.17), and ex-smokers (OR = 1.45, CI = 1.01-2.10). Women were more likely to have 10 or more days of sick leave. Gender differences were reduced mainly when adjusted for type of work (15%) and educational level (7%). Conclusions: The higher occurrence of sick leave among women than among men was partly explained by type of work and educational level. Our results suggest that type of work, a stressful life, and smoking are important targets for health promotion in this study population