KATERINA LYROUDI AND DIMITRIOS ANGELIDIS, UNIVERSITY OF MACEDONIA, GREECE
This paper investigates the magnitude of off-balance-sheet (OBS) activities on the productivity of banks in the Arabic countries. The sample covers nine Arabic countries for the period 1997-2002. We employ the data envelopment analysis (DEA) technique to calculate the Malmquist indices of total factor productivity (TFP) change. This procedure is repeated twice, with and without the inclusion of OBS items. The results revealed that OBS activities do not seem to increase the level of productivity, with the exception of the period 2000-2001. Also, the influence of OBS is not statistically significant for the evaluation of the Arabic banks’ productivity. However, when the OBS items are included in the model the predicted TFP indices are closer to the actual TFP indices for the majority of the cases, since the mean absolute percent error (MAPE) is lower for these cases. Hence, the first model, with the inclusion of OBS items, should be preferred for the evaluation of the banks’ productivity.
Keywords: Arabic countries, bank productivity, data envelopment analysis (DEA), off balance sheet activities, neural network systems