Using full-population register data from Denmark, this study shows that estimates of the economic gradient in mortality depends on the specific measure of economic resources used, where we investigate permanent income, annual income or financial and housing wealth. Our favorite measure is what we call ’Permanent income’, that is the average level of income over a long interval. We find that when using annual income or current wealth, the gradient is overestimated, unless one controls for a number of additional variables, such as education, civil status and initial health. In the last part of the paper, we compare the results from Denmark to results from the UK. Although the countries are very different in terms of inequality, the estimates of the gradient we find are very similar, suggesting that differential levels of resources (including information), rather than inequality itself, determine the gradient in survival and mortality.