Abstract/Summary

This PhD thesis comprises three main chapters, with a unifying theme connecting them. Specifically, the common thread across these chapters is the focus on the redistributive effects of fiscal policy in middle-income countries, with each chapter exploring distinct aspects of this subject matter. Apart from the fact that the study of distributional and fiscal policy matters within middle-income countries is a rather scarce domain of research, the relevance of this subject matter can be elucidated by two crucial points: First, the global trend towards a widening divide between the "rich" and the "poor" is especially acute within these countries. Indeed, middle-income countries stand out as the most unequal among all groups of countries (United Nations, 2015). Second, within middle income countries, key fiscal policy tools, such as taxes and public spending, tend not to be robust, particularly when compared to high income countries. For example, while the tax ratio for advanced economies exceeds 30% of GDP, middle-income countries have a much lower share, typically between 15% and 18% over the past three decades. Meanwhile, the composition and combination of fiscal policy tools matter vis-à-vis their potential distributional effects. In light of the comparatively constrained scale and visibility of key fiscal policy tools within middle income countries, it is therefore reasonable to be interested in discerning how effective fiscal policies would be in recording distributional gains, within this group of countries. This broad discussion about the distributional impacts of fiscal policy within middle income countries has been examined through specific research objectives that are expounded upon in the three chapters of this thesis. In its first chapter, the thesis investigates whether government expenditure reallocations between sectors are equalizing. Specifically, the chapter examines the inequality effects of funding social spending sectors through cuts in other sectoral expenditures, within a panel of 50 middle-income countries over the period 2005–2015, effectively maintaining the same level of overall government expenditure. The chapter identifies the social spending sectors as: social protection sector, health sector, education sector, and the agriculture sector. Meanwhile, the defense sector, transport and communication sector, and "other" sectors are considered for funding the social spending sectors. Further, the chapter evaluates the effect of public spending reallocations on a summary measure of inequality (the Atkinson index) and on three segments of the income distribution (the 10th percentile, the 50th percentile, and the 90th percentile) that reflect three "ideal" income groups. Additionally, sensitivity analyses are conducted by using two other summary measures of inequality (the Gini coefficient and the Theil index), and three other percentiles of the income distribution (the 20th, 40th and 80th percentiles). Further tests are carried out to ascertain if the findings alter when the sample is divided into upper and lower middle-income countries. The second chapter investigates the response of inequality to public expenditure shocks across various time horizons. Specifically, the chapter employs the GMM Panel VAR method in studying a sample of 56 middle-income countries, over the period, 2004-2014. Orthogonalized impulse response functions (IRFs) and forecast error variance decompositions (FEVDs) are used to identify the shocks, which are defined as unanticipated changes in government expenditure. Additionally, the chapter explores the distributional effects of shocks imposed on taxes as well as three social expenditure variables, which are identified as social protection, health care, and education. Further, the chapter assesses how the tax and spending shocks affect those at the bottom (10th percentile), middle (50th percentile), and top of the income distribution (the 90th percentile). The research presented in this chapter represents the first to employ the GMM Panel VAR method towards examining the effects of tax and public spending shocks on the distribution of income within middle-income countries. The third chapter of this thesis employs a Bayesian DSGE model in evaluating the effect of negative commodity shocks on household consumption within Brazil, Russia, India and South Africa (BRIS), on one hand, and Mexico, Indonesia, Nigeria and Turkey (MINT) on the other. Amidst the negative commodity shocks, the effects of fiscal policy shocks on consumption are also considered. In theory, the fiscal policy shocks examined are expansionary, and are accounted for, through transfers, tax and debt shocks. While the shock imposed on tax is negative, that of transfers and debt is positive. Specifically, the chapter evaluates how such fiscal policy shocks affect household spending in the aftermath of negative commodity shocks. It is noteworthy that two distributionally diverse households are considered: those with complete access to financial markets (i.e., Ricardian households) and those without such access (i.e., non-Ricardian households). As a corollary, the chapter examines how both households compare in the aftermath of the shocks, across BRIS and MINT. Further, the chapter compares the changes in domestic and foreign goods consumption following the shocks, thus allowing for the determination of the category of goods that are more impacted by the commodity and fiscal policy shocks. The research presented in this chapter is the first to compare the effects of commodity and fiscal policy shocks on household spending in both BRIS and MINT. As demonstrated thus far, a common thread across all three chapters is that each addresses issues related to fiscal policy, income distribution and middle-income countries. As regards the findings obtained, the first chapter shows that income inequality can be reduced, with all percentiles benefiting, if the education sector is funded by cuts in public expenditure on the transport and communication sector, the defence sector, and "other" sectors. After splitting the data by national income levels, the chapter reveals the equalising role of government spending reallocations in favour of the agricultural sector within the subsample of upper middle-income countries. Similar inequality-reducing roles are also revealed, following reallocations in favour of the social protection and health sectors, within the subsample of the lower middle-income countries. While uncovering the kinds of spending reallocations that are equalizing within middle-income countries, the chapter recommends that, when reallocating away from the relevant financing sectors (i.e., the sectors from which expenditure is being reallocated towards the social spending sectors), policymakers should place greater priority on imprudent expenditures (within the financing sectors), particularly as such inefficient expenditures have remained a major issue to contend with in many emerging countries. Meanwhile, the second chapter of the thesis finds that government and education spending shocks positively impact the low- and middle-income groups, with high-income groups also benefiting from education spending shocks. Meanwhile, social protection shocks often exhibit brief equalising effects; and health spending and tax shocks generally have no detectable effects on income inequality. Further, the third chapter of the thesis shows that households, both Ricardian and non-Ricardian, reduce consumption following negative shocks to commodity output and prices. Among BRIS, Russia witnesses the biggest drop in aggregate consumption, meanwhile, the decline obtained for Nigeria is greater than that of Russia as well as the rest of MINT. Nonetheless, positive shocks on public transfers raise aggregate consumption across BRIS and MINT and play a crucial role in facilitating a redistribution pattern that is associated with a fall in the consumption ratio between Ricardian and non-Ricardian households. Meanwhile, such reductions in the consumption ratio are not observed in the aftermath of negative tax shocks and positive debt shocks.