Browsing by Author "Joseph, Cornel"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    The causal relationship between government expenditure and government revenue in Tanzania, 1966-2010
    (University of Dar es Salaam, 2012) Joseph, Cornel
    This study examines the causal relationship between government expenditure and revenues in Tanzania from 1966 to 2010 for the long run and short run through integrating the Error Correction Model (ECM) into the traditional Granger causality test. The study uses multivariate models, where real GDP and dummies used as the control variables and captures unusual circumstances in the economy respectively. In addition, Impulse Response Functions (IRFs) and Forecast Error Variance Decompositions (FEVDs) have discussed the nexus between government expenditure and revenue. The empirical result shows that there is unidirectional causality running from government expenditure to government revenue in the long run, hence support spend-tax hypothesis. This implies that a budget deficit in Tanzania mainly has been driven by rapid growth in government expenditure. Thus reducing government expenditure should be an optimal solution to curb the budget deficit phenomenon. In addition, the study reveals no causal relationship between government expenditure and government revenue in short run, thus supports the institutional separation hypothesis. Therefore, for the government to reduce the budget deficit in the short run, it should cut public expenditure in unproductive sectors and at the same time ensuring effective utilization of available resources.
  • Thumbnail Image
    Item
    Education, gender and labour market outcomes in Tanzania: evidence from 2014 integrated labour force survey
    (University of Dar es salaam, 2019) Joseph, Cornel
    The focus of this dissertation is on the effect of education and training on labor market outcomes in Tanzania by using 2014 Integrated Labour Force Survey (ILFS) data. First, we examine the effect of education and training on employment mobility in Tanzania by Multinomial Logit Model technique. The result suggests that an individual with secondary or tertiary education is more likely to be in wage employment but is less likely to be in agricultural and non-agricultural self-employment. The findings further suggest that an individual who acquired any training is more likely to be self-employed or engaged in wage employment, but is less likely to be not-working. Second, we estimate the returns to education by using ordinary least square and quantile regression techniques. The result shows existence of convexity in returns to education and the marginal returns to education increases considerably over the quantiles of earning distributions.Lastly, the size of the gender earnings gaps was estimated using Recentered Influence Function decomposition technique. The finding shows the declineof gender earnings gap towards the upper half quantiles and almost varnishing for the highest paid jobs. Thus, results from this study suggest that providing education to females would be a useful investment and good mechanism for realization of the national objective of women empowerment. Likewise, provision of training at different exit points of the general education can improve labour market outcomes. Policies should also give priority to interventions that narrow gender labour market inequality.