Comparing SFA and DEA methods on measuring production efficiency for forest and paper companies

In this study, stochastic frontier analysis (SFA) and data envelope analysis (DEA) are introduced to measure and compare the technical efficiency scores for 79 forest and paper companies. To recognize the environmental effects on the DEA technical efficiency scores, the two-stage DEA was also applied in order to compare any differences in the rank order. The average technical efficiency scores were 0.788 and 0.706 by SFA and DEA, respectively. The highest average efficiency scores, using these two methods, were 0.884 and 0.919 in Japan, respectively. Moreover, the lowest average efficiency scores were 0.608 and 0.396 in Latin America. Although slight differences exist in the scores obtained from these two methods, the highest and lowest relative efficiency ranking for forest and paper companies remain the same. The tests of the rank order correlation were all statistically significant among the SFA, the DEA, and the two-stage DEA methods. The results suggest that policy makers and analysts can be at case when employing these methods to identify the relative efficiency scores for forest and paper companies.