Essays on the dynamics of employment and income distribution

In Chapter 1, I deliver a selective overview of a wide variety of econometric techniques that have been applied over the last two decades to estimate and test the statistical validity of the empirical relationship between output and unemployment, as first postulated by Arthur Okun in 1962. The focus of the chapter is on time-varying and state-dependent methods and it is categorized in: linear and nonlinear parametric methods; non-parametric and semi-parametric methods; and Bayesian state-space methods. The broad consensus in the literature points toward the fact that Okun's law has become less stable than when it was originally conceived, partly due to the existence of different types of nonlinearities affecting the link between output and employment.;Chapter 2 investigates whether the correlation between output and unemployment has become state-dependent, and whether the response is stronger during recessionary periods using a hybrid TSCS approach. Our purpose is to deliver new estimates for Okun's Law for a large number of industrialized and developing countries using a systematic approach which exploits cross-sectional similarities, as well as it allows to have country-specific coefficients. To do so, we used the dynamic specification of Okun's law that includes the effect of asymmetric behavior in contemporaneous output growth (model 1), and on lagged unemployment growth (model 2), given by the state of the business cycle. Our findings point to the existence of nonlinear dynamics between output growth and unemployment, with the effect being stronger during recessions. Additionally, we found that using restricted pooled regressions increased the statistical power of the regressors in countries for which the time series was not representative.;The purpose of chapter 3 is to study the distributive shares of income using a Marxist framework by measuring the rate of surplus value by industry for a sample of 100 countries. This research builds on the work done by Amsden (1981) and extends it to four industries compared to only manufacturing in the original study, over a more recent period of time from 1995-2011. The results support the hypothesis that the rate of surplus value behaves nonlinearly over the development cycle due to, among others, social movements and technological progress. Three out of the four industries present an inverted U-shaped pattern for the rate of surplus value, which suggests that the rate of surplus value is low at low levels of income; it reaches its highest levels at intermediate levels of development, and it tends to be lowest at high levels of income.