Are there any differences in the savings ratio according to income? This question lies at the heart of many public policy questions. Are taxes on consumption regressive? What is the effect of an increase in income tax on high incomes? Should pension savings be subsidised? Observation of savings ratios according to current incomes reveals a positive link. Saving may, however, be a long term decision which does not depend solely on current income. If households anticipate a drop in their income, they tend to “smooth” their consumption and, in the event of any exceptionally high income, they may well adjust their savings upwards. The true savings behaviour must therefore be studied more according to permanent income, that being defined as the income which corresponds to the updated flow of resources that the household expects to get over a long period, corrected for transient fluctuations. On the basis of the 2010 Wealth survey, we study the link between savings and the two concepts of income - current and permanent - using different approaches to estimate income. Previously, the five different approaches employed have never been used in conjunction with each other. This makes it possible to compare the findings obtained and their sensitivity to the estimation methods. All our results suggest that the savings ratio is significantly higher in those households with the highest incomes, irrespective of the definition of that income.