This thesis was informed by using three existing datasets. The main dataset (RAFiP) was collected as part of a randomised controlled trial conducted in Ghana in 2015 and 2016. Baseline data was collected on 1,441 respondents while endline data was collected on 1,415 respondents. This data covered sections on demographic information, financial literacy, household consumption, asset accumulation, financial inclusion, health expenditure and others. The other two datasets are the sixth and seventh rounds of the Ghana Living Standards Survey (GLSS) that was collected by the Ghana Statistical Service. GLSS6, which was collected in 2012/13, sampled 16,772 households while GLSS7 (collected in 2016/17) had a sample size of 14,009 households. The GLSS surveys cover sections on demography, housing conditions, employment, education, water and sanitation, health, access to financial institutions and insurance services, remittance and household assets, poverty, disability, migration, agriculture, non-farm activities and governance, among others.

Firms of differing sizes make FWAs available to employees, with varying performance outcomes. Research on the financial outcomes of FWAs is sparse and tends to focus on large firms. This study investigates the associations between FWAs and return on labour (ROL) as well as the relevance of these associations to small, medium and large firms, using a sample of 3244 employees working in 602 businesses. The findings show negative associations between flexible leave as FWA and ROL for all firms. Job-sharing has financial value for firms with 100 or more workers, with the majority being females but it is not feasible in small firms due to limited employee numbers. Flexible work hours pay off for firms with up to 99 employees but the financial outcomes become negative thereafter, requiring closer monitoring in larger firms. The findings indicate that firm size is relevant to FWA regulations and negotiations with implications for employers, employees and policymakers.

Purpose – The purpose of this paper is to determine the effect of employment security on moonlighting in Ghana as a means to inform policy on enforcing issues of employment security.
Design/methodology/approach – The paper followed the work of Shishko and Rostker (1976) by applying their approach to the sixth round of the Ghana Living Standards Survey data. The paper created an employment index from four variables and used the ordered logit as the main estimation technique.
Findings – The findings indicated that, as a person with a single job becomes more secured in employment, the likelihood of moonlighting is decreased by 0.03, while for those with two and three or more jobs, the likelihood of moonlighting increased by 0.0297 and 0.0008, respectively, with increasing employment security. This implies that, workers can be made to stick to single jobs by providing them with higher levels of employment security, but once they take on two or more jobs, providing them with employment security will engender the tendency to increase their moonlighting behavior the more.
Originality/value – With the current harsh economic conditions in the country and the urgent need for multiple jobs (moonlighting) as a risk coping mechanism, little has been done on the role employment security plays as a catalyst or otherwise. This paper fills the gap by employing a comprehensive index on employment security in the case of Ghana.