In this paper, we use a new source of linked Australian census, tax, social security and Medicare data to analyse the characteristics of those who were at the very top of the income distribution in 2011. The Basic Longitudinal Extract 2011 (BLE2011), from the Multi-Agency Data Integration Project (MADIP) overcomes a number of limitations of previous datasets. In addition to providing tax data for a very large proportion of the adult population, it combines census data linked at the individual level. Importantly, it has a household identifier, which allows us to calculate the distribution of equivalised household income, as well as the distribution of individual income. We show that there is quite substantial movement in and out of the top of the income distribution, depending on whether we use individual or household data. Furthermore, despite some assumptions to the contrary in the popular discourse, a much higher proportion of people at the top of the equivalised household taxable income distribution are professionals, as opposed to managers. Finally, although receipt of social security is quite low at the very top of the income distribution, a nonnegligible number of people in the top 2% of the income distribution still received some form of payment or allowance; the most common payments were Carer Allowance, the Seniors Health Card, the Age Pension, and Family Tax Benefit Part B and Part A (in that order).