Accurate national‐scale economic assessments of flood risk should, first, allow policy and resource allocation to be determined sensibly by relating both to the scale of that risk and, secondly, provide a benchmark against which any possible future changes in risk can be assessed. This paper critiques the UK's National Flood Risk Assessment (NAFRA), its development for England and Wales over a decade, and its evaluation of economic fluvial and coastal flood risk. We find, when comparing the results with other data on flooding and the damage caused over 20 +  years (including insurance claims), that NAFRA appears to overestimate the economic risk by between four‐ and five‐fold (i.e. at c.£1.1bn p.a., as opposed to a central estimate here of £0.25bn p.a.). The flood problem faced is still serious, and may be increasing, but we suggest that an exaggeration has become embedded. The assessment models and their data need further enhancement if this situation is to be corrected and a realistic assessment made of the scale of the current fluvial and coastal flood risk. Only then will the results be adequately credible and hence useful.