Increasing global interaction between economies over the last few decades has led to growing interest on the implications of globalization. Of particular interest has been the distributional impact of globalization and whether this has been equity enhancing. Contributing to this debate, Chotikapanich et al. have previously estimated global and regional inequalities for 1993 and 2000. The current paper presents estimates of global inequality for 2005, making use of purchasing power parity (PPP) data from the 2005 International Comparison Program (ICP) round. A critical element in the estimation of regional and global inequality is the data on PPPs and real incomes. Another objective of the paper is to examine the sensitivity of the inequality results to the use of alternative sets of real incomes using different sets of PPP data to convert per capita income data into a common currency unit. The paper also compares levels and trends of global inequality measures based on real per capita income measures from the national accounts and those obtained using per capita expenditure data from household expenditure surveys. The main finding of the paper is that the populist view that globalization has increase inequality does not hold when inequality is measured at the global level. Between 1993 and 2005, inequality has consistently declined as measured both by the Gini coefficient and the Theil index. This decline in global inequality was the result of a decline in inequality between countries. The estimates do suggest that there is an overall upward trend in inequality within countries, lending some support for protestors against globalization. Another finding is that the levels of global inequality are indeed sensitive to the choice of PPPs. However, the downward trend in global inequality is consistently evident across different choices of PPPs and real incomes. We find similar trends in inequality and in the within and between components of inequality when inequality measures are based on national accounts real per capita incomes or on the survey-based real per capita expenditures.