This study measured the inefficiency of the banking industry over a ten-year period from 2004 to 2013 using directional distance functions (sequential directional distance function and contemporaneous directional distance function), and based on these measurements, calculated the sequential Malmquist-Luenberger productivity index and the Chung et al. (1997) Malmquist-Luenberger productivity index to analyze productivity changes in the banking industry. The purpose of this study was to compare and analyze the two productivity index models and use the sequential Malmquist-Luenberger productivity index—which more realistically explains the situation of Korea's banking industry—to compare and analyze efficiency and productivity changes in the banking industry before and after the global financial crisis, as well as efficiency and productivity changes between nationwide commercial banks and regional banks. The analytical results of this study are as follows. First, inefficiency in Korea's banking industry has generally increased since the 2008 financial crisis. Second, productivity in Korea's banking industry has shown an overall downward trend, which is interpreted as a result influenced by the failure of technical efficiency to improve and its subsequent decline, rather than being attributable to technological change. Third, both nationwide commercial banks and regional banks have experienced overall declines in productivity, with regional banks declining more substantially than nationwide commercial banks. In conclusion, this study suggests that productivity changes measured based on the contemporaneous frontier are primarily influenced by technological change, whereas productivity changes measured based on the sequential frontier are attributable to technical efficiency, implying that the interpretation of productivity change factors can differ depending on which productivity index model is applied. However, regarding the phenomenon where best practices move into the interior of the previous frontier during the frontier construction process, treating this as technological regression is arguably inappropriate if one considers the long-term nature of technology and the accumulation of past information through learning effects that cannot be ignored; rather, it is more appropriate to interpret this as a decline in technical efficiency. This study has a limitation in that while risk should be considered when measuring banking industry efficiency, only non-performing loans were considered, and liquidity risk, capital structure risk, and other risks were not controlled for in the research model. Therefore, to address this issue, additional research is needed that applies stochastic frontier analysis to decompose banking industry inefficiency into managerial inefficiency and environmental factor inefficiency, and then analyzes inefficiency after controlling for environmental factors.