Essays on Self-Employment

This thesis consists of three chapters. All of the three chapters are interlinked and revolve around self-employment and its impact on inequalities and households welfare. This dissertation aims to further our knowledge of self-employment to inform the necessary redesigning of labour market policies.

Chapter one of this thesis uses data from the Bank of Italy's Survey on Household Income and Wealth (SHIW) to investigate whether the type of employment, that is, whether a person is self-employed or works for others (paid employed), matters for consumption insurance. Our Ordinary Least Square (OLS) and Two-Stage Least Square (2SLS) empirical results suggest that households headed by self-employed are better able to insure/protect their consumption from idiosyncratic income shocks than households headed by paid employees. Furthermore, the non-durable consumption of self-employed is better insured than durable consumption.

Chapter two investigates whether the trends in aggregate income and consumption inequality vary across payroll and self-employed households using SHIWdata. This analysis further seeks to analyse, to what extent changes in aggregate consumption and income inequality can be explained by changes in its permanent and transitory components.Our GMM results suggest that there are some considerable differences in the consumption/income inequality and its permanent and transitory components across self-employed and payroll households. For self-employed households, both income and consumption in-equality is higher, the increase in total income and consumption inequality is mainly driven by an increase in its transitory component. Empirical results from other measuresof inequality (the Gini coefficient, the variance of the log, 90th/10th, and 50th/10th percentile ratios) con?rm that income inequality is higher than consumption  inequality for all groups and more pronounced in the case of self-employed households.

The third chapter addresses three different questions using data from British Household Panel Survey (BHPS) and the United Kingdom Household Longitudinal Study (UKHLS). First, to investigate whether the premium/penalty to self-employment has changed over time?. We use different estimation techniques such as ?fixed effects, pooled OLS, and random effects to estimate trends in wage premiums/penalties over time.

The second question is, what contributes to the raw wage gap between paid and self-employed?. The Oaxaca Blinder decomposition method is used to answer this question and fi?nd the contribution of explained and unexplained components to the raw wage gap. The third question is whether mobility from self-employment to paid employment reduces the wages of workers?. We use a novel estimation technique, Quantile Regression for Panel Data (QREGPD) to answer our fi?nal question. Our fi?xed effects regression results suggest that the earning penalty to self-employment has decreased over time, but it is still negative. The Blinder-Oaxaca decomposition results reveal that the raw wage gap between paid and self-employed has decreased over time due to a reduction in the unexplained component. QREGPD results indicate that mobility from self-employment to paid employment entails a positive premium to workers at the lower quantiles and as we move up to the 90th quantiles the wage premium to mobility disappears.