The objective of the thesis is to develop a method to improve current mortality based measures of wellbeing and inequality. The basic assumption made in the thesis is that mortality risks can be divided into avoidable and unavoidable mortality risks. Unavoidable mortality risks are beyond the control of human given the available resources and technology and therefore should be excluded from any measurements of wellbeing and inequality. The method developed in the thesis involves quantifying unavoidable mortality risks using year 2000 life tables data for 191 countries resulting in a reference country mortality distribution. The thesis then makes use of the constructed reference distribution to improve a number of existing mortality based measures, including life expectancy, potential years of life lost, and age at death. The new measures of welfare are empirically implemented, compared and contrasted with more commonly measures such as per capita income, Human Development Index and income inequality measures. The improved measures will be used to examine gender inequality and within country inequality for the 191 countries. Finally, it will be used to investigate racial inequality between Maoris and non-Maoris in New Zealand and between White and Black Americans in the United States. Based on the computed improved mortality measures, we find that controlling for unavoidable mortality risk can make significant difference when measuring and wellbeing inequality.