1. But the literature suggests that bank size matters. Some studies find increased productivity growth (in terms of costs or profits) in the early 1980s for large banks (due to shifts in the best-practice frontier) but not for small banks. For example, Humphrey and Pulley (1997) found that profits of larger banks in the sample (with assets over $500 million) increased by 12 percent between the 1977-1981 period and the 1981-1984 period. Decomposing this change, they found that it results form a shift in the profit function and changes in business condition, particularly deposit deregulation. Only business conditions accounted for the risk in large banks' profits form 1981-1984 to 1985-1998. For smaller banks (assets of $100 million to $500 million), there was little increase in profits between 1977-1981 and 1971-1984. Wheelock and Wilson (1999) used linear programming techniques (DEA) and decomposed the change in productivity into the change in efficiency and the shift in efficient frontier. They found that banks on the frontier improved over the period 1984-1993 but productivity declined, on average, during this period because of reductions in efficiency;There are fewer studies of productivity in banking. Using data on banks from the late 1970s and 1980s, most studies find negative cost productivity growth, on the order of ?1 percent per year. Using panel data on 661 top-tier bank holding companies continuously in existence during,1991
2. look at both cost and profit productivity, where productivity is measured as a combination of technological change and changes in inefficiency, holding constant the References Akhavein;Berger;Review of Industrial Organization,1997