Accounting for environmental factors, bias and negative numbers in efficiency estimation: a bootstrapping application to the Hong Kong banking sector Maximilian Hall Karligash Glass Richard Simper 2134/6003 https://repository.lboro.ac.uk/articles/preprint/Accounting_for_environmental_factors_bias_and_negative_numbers_in_efficiency_estimation_a_bootstrapping_application_to_the_Hong_Kong_banking_sector/9492944 This paper examines the evolution of Hong Kong’s banking industry’s technical efficiency, and its macroeconomic determinants, during the period 2000-2006 through the prism of two alternative approaches to efficiency estimation, namely the intermediation and production approaches. Using a modified (Sharp, Meng and Liu, 2006) slacks-based model (Tone, 2001), and purging the efficiency estimates for random errors (Simar and Zelenyuk, 2007) , we firstly analyse the trends in bank efficiency. We then identify the ‘environmental’ factors that significantly affect the efficiency scores using an adaptation (Kenjegalieva et al. 2009) of the truncated regression approach suggested by Simar and Wilson. 2007). The first part of the analysis reveals that the Hong Kong banking industry suffered a severe downturn in estimated technical efficiency during 2001. It subsequently recovered, posting average efficiency scores of 92 per cent and 85 percent under the intermediation and production approaches respectively by the end of 2006. As for the sub-group analysis, commercial banks are, on average, shown to be the most efficient operators, while the investment bank group are shown to be the least efficient. Finally, with respect to the truncated regression analysis, the results suggest that smaller banks are more efficient than their larger counterparts, although larger banks are still able to enjoy gains from scale economies and benefit from the export of financial services. Moreover, private housing rent and the net export of goods and services are found to be negatively correlated with bank efficiency, while private consumption is shown to be positively correlated. 2010-03-15 12:22:25 Hong Kong banks DEA Slacks Environmental factors Negative numbers Bias Economics not elsewhere classified