Non-linear effects of credit risk on efficiency and productivity of Iranian banks: threshold panel regression and Malmquist approach

The economic conditions of the country have caused borrowers to have problems in repaying their loans, and as a result, the non-performing loans of banks have expanded, which means an increase in credit risk. The purpose of this article is to evaluate the non-linear effects of credit risk on the efficiency and productivity of 15 stock market banks in the period of 2014-2022.The Malmquist index and technical and efficiency changes were calculated to measure efficiency and productivity, that the efficiency change also was decomposed into net efficiency and scale efficiency. The findings showed that the productivity of banks increased by 6.3% annually; which its 1.9% was due to technical change and 4.3% due to efficiency improvement. In second section, the effect of credit risk on efficiency and productivity was estimated with an emphasis on bank size using the threshold panel regression method. The results of model (1) showed that the effect of credit risk on the banks productivity is in three forms according to the size of the bank as a threshold variable: when the size of the bank is less than 25 or more than 27, a one percent increase in credit risk will decrease productivity by 0.06 or 0.04 percent respectively; and when the bank size is between these two values, this effect is equivalent to 0.11 percent. The findings of model (2) also showed that when the size of the bank is less than 14 or more than 16, a one percent increase in credit risk causes a 0.03 or 0.04 percent decrease in bank efficiency, respectively.