Early Warning Models for Banking Supervision in Romania

Within this document we suggest an early warning system for the Romanian banking sector, to supplement the standardized CAAMPL rating system employed by the National Bank of Romania for evaluating the local credit institutions. We make an effort to discover the determinants for downgrades as well as for a bank to have a weak overall position, to determine the respective probabilities and to be able to perform rating predictions. Having one of these objective, we develop 2 models with binary dependent variables and one ordered logistic model which accounts for all possible future ratings. One outcome is that indicators for present position, market share, profitability and assets quality determine rating downgrades, whereas capital adequacy, liquidity and macroeconomic environment aren’t displayed in the model. Banks which possess a weak overall position in one year could be forecasted using also indicators for current position, market share, profitability and assets quality, as well as, in cases like this, capital adequacy and macroeconomic environment, the latter only for the binary dependent variable model, leaving liquidity indicators out again. Depending on the ordered logistic model’s capacity for rating prediction, we approximated one year horizon scores and ratings for each bank and we aggregated these results for predicting a measure of assessing the local banking sector in general…

Banking is probably the most intensively supervised industries world-wide because of the high-impact of bank failures on economic activity. Financial stability, a multitude of markets, infrastructure and even people’s personal safety and comfort rely on the credit mechanism and the soundness of the banking sector. For that reason, around the globe, governments grant authority to financial supervisory bodies and put them in charge with the regulation, authorization and supervision of the financial institutions, to be able to limit the potential risks they undertake and the negative effect they may have on other economic sectors…

Source: Bucharest University of Economics,

Early Warning Models for Banking Supervision in Romania