Financial ratios are used for all kinds of purposes. These include the assessment of the ability of a firm to pay its debts, the evaluation of business and managerial success and even the statutory regulation of a firm's performance. Not surprisingly they become norms and actually affect performance.' The traditional textbooks of financial analysis also emphasise the need for a firm to use industry-wide averages as targets (Foulke, 1968), and there is evidence that firms do adjust their financial ratios to such targets.* Whittington (1980) identified two principal uses of financial ratios. The traditional, normative use of the measurement of a firm's ratio compared with a standard, and the positive use in estimating empirical relationships, usually for predictive purposes. The former dates back to the late nineteenth century and the increase in US bank credit given as a result of the Civil War when current and non-current items were segregated and the ratio of current assets to current liabilities was developed (Horrigan, 1968;and Dev, 1974). From then the use of ratios both for credit purposes and managerial analysis, focusing on profitability measures soon began. Around 1919 the du Pont Company began to use its famous ratio 'triangle' system to evaluate its operating results, underpinning the modern interfirm comparison scheme introduced in the UK by the British Institute of Management and the British Productivity Council in 1959.The positive use of financial ratios has been of two types: by accountants and analysts to forecast future financial variables, e. g. estimated future profit by multiplying predicted sales by the profit margin (the profit/sales ratio), and, more recently, by researchers in statistical models for mainly predictive purposes such as corporate failure, credit rating, the assessment of risk, and the testing of economic hypotheses in which inputs are financial ratios. These will be reviewed in the section on predictive studies.The reason ratios are used, as opposed to absolute values, is a mathematical one, and is basically in order to facilitate comparison by adjusting for size. However, this assumes that ratios possess the appropriate statistical properties for handling and summarising the data. Also, the statistical models assume,