It has been suggested that dirty surplus accounting (violation of the clean surplus relationship (CSR)) may result in mismeasurement of performance and value, and that cross-country variation in dirty surplus accounting may cause particular problems for international comparisons. Using articulated data that are largely hand-collected, we evaluate the potential impact of dirty surplus accounting in France, Germany, the UK and the US for the period 1993-2001. First, we report summary statistics on dirty surplus accounting flows. These indicate that distributions of dirty surplus flows are often not centred on zero, and that there is significant cross-country variation in such flows. Then, we use a measure of multi-period abnormal performance to document the bias and inaccuracy, and cross-country variation therein, that would have arisen from omitting dirty surplus flows in measuring performance. Where significant bias and cross-country variation therein arise, they are largely caused by omission of goodwill-related flows, which regulators are eliminating as a dirty surplus item. In contrast, all classes of dirty surplus flow contribute to significant cross-country variation in inaccuracy. Finally, we address the issue of dirty surplus flows from the valuation perspective. We use the residual income valuation model, which relies partly on CSR, to test whether perfect-foresight forecasts of dirty surplus accounting flows explain beginning-of-interval market-to-book ratios after controlling for other inputs to the valuation model. We find little evidence to suggest that omission of dirty surplus flows from residual income value estimates would have caused systematic valuation errors in the period and countries examined.