Despite the importance of the accounting industry, efficiency evaluations are analyzed in small numbers, and four accounting firms have a very high share of the industry. The transparency of accounting has been intensified since the accounting fraud of domestic Daewoo shipbuilding in Kore, Foreign Enron and WorldCom in the United States. And Importance of appropriate assessment has been highlighted after the increase of lawsuits in case of the occurrence of the bad accounting and the change of accounting industry due to the 4th industrial revolution.
The analysis of efficiency uses data envelope analysis(DEA: CCR and BCC) method which determines the most advantageous weight to oneself. This method is easy to calculate the efficiency and has an advantage that the improvement can be easily obtained. And the advantage of this method is that it can be used to evaluate the degree of efficiency by using the industry average or median value. The disadvantage of this method is that it has the most advantageous weight in the calculation of efficiency, so it is sometimes possible to increase the proportion of some inputs and rarely use of some inputs.
For example, in accounting firms, when using the number of accountants and the number of clerks as input variables, it is possible that efficiency is achieved by using only 99% of the clerks and only 1% of the accountants. These specific corporations have features that are not suitable as a benchmarking tool despite efficiency. Therefore, since there is no standard for eliminating these accounting corporations, it is difficult to determine anomaly by simply not being referenced.
Cross-efficiency, which is a peer evaluation reviewed the problem presented in the standard DEA. This method applies the weight of each DMU using the multiplier model to the corresponding DMU, and averages them. Although the cross-efficiency is meaningful, the proportion of each DMU is not unique according to the application of the multiplier model, so that different results are obtained depending on the use of the program optimization method. Commercial software also shows each different weight. The game cross-efficiency is a way of solving this non-unique weight. This is a way to apply the game to the cross-efficiency, optimizing oneself competitively for other DMUs, and finds a unique weight, and converges both the reciprocal efficiency and the aggressive efficiency into a single value.
There is a Common set of Weights, which emerges as a common evaluation. This method has attracted a lot of attention because DEA and cross-efficiency analysis have the problem of using a proportion that is advantageous to oneself. When evaluating a DMU, evaluating the superiority of the test with other test papers itself is not meaningful in ranking. Therefore, Kao, Hung (2005) suggests a common method for analyzing the efficiency of the industry.
According to the efficiency calculated by the above three methods, it is problematic to interpret the staggered ranking in the case 10th ranking in one criteria and 100th ranking in the other criteria. Shannon entropy method has been suggested as a method to obtain comprehensive efficiency by objectively evaluating the importance of each method and presenting the weight.
The main result of this study is that the efficiency by entropy is suitable for deriving the criteria that can be used in the market. For example, Suhlin Accounting Corporation was the most efficient in terms of CCR, Maverick, the general leader, but this corporation was ranked 23rd in Entropy, which is the result of the above self evaluation, peer evaluation and common evaluation. HanYoung accounting corporation, which belongs to the 4 largest accounting corporations, is ranked 24th to 9th, showed that appropriate evaluation is being made.
In DEA''s analysis, the inefficiency of the accounting industry is found to be the operational efficiency, and the context-dependent DEA is divided into six tiers. The main conclusion of this paper is that it is more realistic to evaluate the efficiency of economic DMUs by combining the standard DEA with peer evaluation and common evaluation.