his paper examines trends in the wealth and incomes of older Australians over the 11 years from 1985-86 to 1996-97. The primary data sources for this analysis are the ABS 1986 Income Survey and 1997-98 Survey of Income and Housing Costs ‘confidentialised’ unit record files. These surveys provide very comprehensive national snapshots of the income, socio-demographic and other characteristics of a representative sample of Australians. The ABS defines income as ‘cash receipts that are regular and recurring’ and includes income sources such as wages and salaries, profit and loss from own business, property income, government cash transfers and private cash transfers (such as superannuation and child support) are all included. Receipts that are excluded from income because they are not regular and recurring include inheritances, maturity payments on life insurance policies and capital gains and losses. Imputed rent is not calculated for home owners. Annual cash income is the income measure used. While income is a flow, wealth is a stock. Both are important when considering the economic position of older Australians. Two older Australians with similar incomes might be in very different positions if one owns no assets but the other has substantial assets to draw upon if and when they are in need. The ABS surveys provide details of only one type of wealth — the respondent’s estimate of the value of their home and the amount of mortgage outstanding on that home. The surveys do not directly provide estimates of the value of other wealth sources, but instead report the income received from such wealth sources (such as interest, rent and dividends). This information has been used by NATSEM to impute estimated wealth holdings for each of the families contained within the survey. For 1986 this utilised the methodology mapped out by Baekgaard (1998) with some later minor adjustments. For 1996-97 the methodology is described in Kelly (2001). The wealth sources included in this study comprise own home, shares, cash, investment properties, own business and accumulated value of superannuation. Other sources, such as consumer durables and cars, are not included. The discounted future value of the government age pension is also not included within our estimates of private wealth holdings. It must be emphasised that imputing the estimated value of wealth holdings using such capitation methods, whilst a technique frequently employed by researchers in this field (Dilnot, 1990), is nonetheless subject to some degree of