We present evidence that scores on the Montgomery, Hamilton and Beck depression scales are strongly intercorrelated and can be equated to each other. These scales were administered on 107 occasions to patients with major depression. The correlations between scores on each scale were much higher than has been previously reported and, furthermore, the relationships were always well described by a linear function. The close correlations could not be explained by rater biases. We were able to provide validation data for two of the rating scales, demonstrating that our simple models generalize well to a novel dataset. The knowledge that there is a robust relationship between the three scales has practical value for both clinicians and researchers.