Introducing the Morningstar Solvency Score, A Bankruptcy Prediction Metric

This paper examines the performance of the Morningstar Solvency Score, Morningstar's new accounting-ratio based metric for predicting bankruptcy, in comparison to the Altman Z-Score and Distance to Default models. Specifically we tested the following: 1. The ordinal ability of each model to distinguish companies most likely to file for bankruptcy from those least likely to file for bankruptcy as measured by the Accuracy Ratio 2. The cardinal ability of each model to predict bankruptcy as measured by the bankruptcy rate of healthy-scored companies and the average rating before bankruptcy 3. The decay of the ordinal performance of each model over time as measured by the cumulative percentage change in Ordinal Score 4. The stability of the ratings of each model as measured by the Weighted Average Drift Distance We found that the Morningstar Solvency Score has superior ordinal and cardinal bankruptcy prediction power than our comparison models within a one year time horizon. It also exhibited a lower drift rate than Distance to Default, although it had less signal durability. The Solvency Score also exhibits low correlation to the Distance to Default and TLTA models, suggesting that it incorporates enough unique information to be useful in combination with the other models. Our testing universe consisted of both manufacturing and non-manufacturing companies, although we did exclude financial firms. We are cognizant that the Z-Score was never intended to gauge the financial health of non-manufacturing companies. However, it is often used to gauge the health of non-manufacturing companies in practice. Therefore we found it more relevant to test all of our models against a universe inclusive of non-manufacturing companies.