Applying Benford’s law to detect accounting data manipulation in the banking industry.

Abstract

In this paper, we take a glimpse at the dark side of bank accounting statements by using a mathematical law which was established by Benford in 1938 to detect data manipulation. We shed the spotlight on the healthy, failed, and bailed out banks in the global financial crisis and test whether a set of balance sheet and income statement variables which are used by regulators to rate the performance and soundness of banks were manipulated in the years prior to and also during the crisis. We find that banks utilise loan loss provisions to manipulate earnings and income upwards throughout the examined periods. Together with loan loss provisions, problem banks resort to a downward manipulation of allowance for loan losses and non-performing loans with the purpose to tamper earnings upwards. We also provide evidence that manipulation is more prevalent in problem banks, which manage income and earnings to conceal their financial difficulties. Moreover, manipulation is found to be strengthened in the crisis period; it is also expanded to affect regulatory capital. Overall, banks utilise data manipulation without yet resorting to eye-catching manipulation strategies that may attract the scrutiny by regulators. Benford’s Law appears to be a suitable tool for assessing the quality of accounting information and for discovering irregularities in bank accounting data.