Official estimates of personal income and its inequality, based on the Household Expenditure Surveys, suffer from inaccurate definition of income and inappropriate procedure for the estimation of inequality. This paper re-estimates personal income and expenditure and their components from the Household Expenditure Surveys of 1991/92 and 1995/96, estimates Gini ratios of income and expenditure distributions and corresponding concentration ratios of the distributions of their components. These results show that the level of inequality in Bangladesh is lower than the official estimates suggest while the rate of increase in inequality during the period under review has been greater than is shown by official estimates. The rising inequality has largely been due to the rising share in income of certain components that are disequalizing (i.e., disproportionately concentrated among the higher income groups) as well as a rise in the extent of their disequalizing effect. For Bangladesh as a whole a good part of increased inequality has been due to the sharp increase in inequality between urban and rural areas. The paper concludes by considering the implications of the findings for policies for poverty-reducing growth in Bangladesh.