| 摘要 | ��
Research Purpose
Recently, it has become increasingly important to evaluate the managerial performance of energy transportation and distribution companies. This indicates that the type of control of energy corporations is changing from the rate-of-return regulation to incentive regulation. As a result, the regulator becomes more interested in providing incentives to regulated corporations so efficiency can be achieved. Along with this, the regulator, who is subject to performing a standard incentive program through benchmarking, must be informed about the regulated corporation's productivity.
Structurally, domestic retail gas distributors are inefficient because of regional monopoly and a guaranteed return on investment under the rate of return regulation. An assessment of the retail gas distributors' efficiency, however, has never been made thus far. As a result, no efficiency improvement efforts have been made on the basis of this assessment. In the short-term perspective, an efficiency evaluation of the retail gas sector through benchmarking can provide a measure of control to minimize or prevent the occurrence of unreasonable profits. In the long term, this evaluation can present realistic targets to improve efficiency when the retail gas sector begins to restructure and introduce incentive regulations.
It should be noted, however, that productive efficiency can be affected according to the characteristics of the supply area and the demand type in which each distributor is placed. Hence, a fair and objective assessment of productive efficiency must be subject to the development of proper techniques that effectively reflect the operating differences that exist among distributors.
Based on this background, this research tries to develop an efficiency estimation method that can reflect the different operating environments of gas distributors, and that can measure their productive efficiency.
Summary
First of all, in the chapter for input and output analysis of the gas retail business, supply environments and financial performance are analyzed for 32 city gas distributors, and a series of partial factor productivity was estimated. Gas delivery per customer has been stagnant between 1997 and 2003 in the Seoul metropolitan area, but it has been increasing rapidly in other large and small provincial cities. On the other hand, the number of customers per length of network has increasingly stabilized in the Seoul metropolitan area, but this has remained stagnant in other large and small provincial cities. The earning ratio to total assets, as an indicator of financial performance, does not show a big difference in terms of area, but the earning ratio to equity does. Gas delivery per employee, as an indicator of labor productivity, shows a difference of about 9.6 times at the highest among distributors, and the number of customers per employee varies by about 6.4 times at the highest. The gross capital investment efficiency, as an indicator of capital productivity, shows no big difference among distributors, except for companies that have experienced some changes in capital structure. The number of customers per length of network differs by a maximum of about 2.4 times among distributors.
In the chapter analyzing the evaluation methods of productive efficiency, the method that is effectively applicable to this report in theory and in practice has been selected based on an analysis of existing evaluation methods and practices that were theoretically developed and executed in other countries. Theoretically, the characteristics, merits and demerits of three evaluation methods - the index approach, parametric approach, and non-parametric approach - were compared. Ultimately, a DEA analysis, which is believed to be suitable to the characteristics of the domestic gas industry's environment, was chosen as a practical method for this report. DEA analysis using physically measured input and output data has several merits, such that it can divide productive efficiency into pure technical efficiency and scale efficiency. It can also reflect environmental factors in various ways, and can be utilized to measure changes in efficiency over time.
Practical ways of measuring efficiency indices were determined by investigating the practical use of the theoretical methodology, the composition of the analysis sample, the composition of the analysis model, the methods to reflect different operating environments, and the results of efficiency measurements that appeared in previously conducted researches. DEA analysis seems to have been widely used as an efficiency evaluation method. The way of adopting input and output data slightly differs from each other, but the length of the network and the number of employees were used as inputs, and gas delivery volume and the number of customers were used as outputs in many cases. Although there are substantial differences in the application of environmental factors among research reports, climate and population density will be representatively utilized in case environmental factors are applied.
The productive efficiency of the domestic gas business has been evaluated by several DEA analysis models established through this kind of methodology analysis. Efficiency evaluation has been performed for production activity between 1999 and 2003. The first evaluation targeted all city gas companies, while the second evaluation was confined to city gas companies supplying LNG.
Two methods of applying the measured efficiency index in incentive regulation were considered. One is to include an adjustment element of the X-factor in the price cap formula. The other is to reflect all the efficiency differences in the initial value of the regulated price cap.
The former method is more practical than the latter because it allows the firm to gradually improve its efficiency, taking into consideration its capacity for efficiency improvement. The firm's potential ability to improve its efficiency, provided there is no structural change, can be compared with its past performance. Having compared a measured efficiency against changes in efficiency index over time, most inefficient distributors appeared to have shown lower increase in efficiency than the efficient distributors.
Despite the efficiency measurement reported here, more credible efficiency evaluation is needed so that the results can be used to set up an effective policy of incentive regulation in the future. Additional research tasks include the development of methods reflecting environmental factors, the establishment of a frontier for efficiency from the sample that includes other countries, a sensitivity analysis adopting other measurement methods, an on-the-spot survey to improve data credibility, and an evaluation of economic efficiency using financial data.
Research Results and Policy Suggestions
The results of the efficiency evaluation can be summarized as follows:
First, the level of productive efficiency measured in the pure technical efficiency index is, on average, between 87% and 96%. There is a great possibility for efficiency improvements even though these differ from company to company. The productive efficiency of LNG-based city gas distributors is, on average, 87%. The efficiency difference among distributors is significant, as the lowest efficiency is 67%. The inclusion of operating environments in the model, however, slightly increased the efficiency scores. In particular, with the customer density variable, the level of efficiency rose by an average of 10% point, and the lowest efficiency also jumped up to 83%, significantly reducing the gap in efficiency between the inefficient firm and the frontier firm. It should be noted, however, that the direct inclusion of environmental variables in the DEA model as inputs and/or outputs can cause the swelling of efficiency scores. Thus, the above results should be interpreted carefully.
Second, operating environments, such as climate, customer density, demand structure, etc., significantly influence the measured efficiency difference among distributors. Since these factors are external regardless of management ability or business strategy, their effects on the efficiency scores should be eliminated to obtain objective results.
Third, the efficiency increase of domestic gas retail distributors in the last five years was evaluated to be weak (annual average of 1.2%), and this efficiency increase resulted mostly from the increase in scale efficiency. Technology improvement appeared to have remained stagnant over the last five years, and the annual average increase of pure technical efficiency during the same period was low (0.1%).
Fourth, it is highly possible for city gas distributors to achieve 100% scale efficiency when the number of customers is between 500,000 and 1,000,000. Distributors that supply 500,000 customers or less are under increasing return to scale and distributors that supply over 500,000 customers are under decreasing return to scale.
The above results provide several policy implications toward incentive regulation as follows:
First, operating environments should be carefully considered in evaluating the efficiency of city gas distributors. To do this, the efficiency difference caused by the difference in the operating environment has to be estimated objectively. Since the levels of effect of each environment on the individual firm, however, may differ from each other, the selection of the environmental factors under consideration in evaluating efficiency needs to be made via a feedback process with city gas distributors.
Second, the low increasing trend in distributor's productive efficiency and the stagnancy in pure technical efficiency indicate that the gas retail business stands a small chance of achieving technology development. Policy-makers need to consider this fact when they establish policies to implement incentive elements in the gas retail industry. This is because adopting the price cap system, which is a kind of incentive regulation in the gas industry, may increase the normal price without any improvement in efficiency in case that the industry has only a small chance of achieving technology development.
Third, when setting up the regulation that induces efficiency improvements, the regulatory regime that reduces the efficiency difference among distributors is more desirable than the one that gives efficiency improvement targets to the entire industry. This policy suggestion is based on the fact that the increase in efficiency over time has been low, but the efficiency difference among distributors has been significantly high.
Fourth, when considering the fact that low-efficiency distributors have a low increasing rate of efficiency, the elimination of inefficiency in the gas retail business should be approached from the long-term perspective. This is because too strict targets to improve efficiency may cause insufficient compensation to gas distributors, and may make them difficult to raise investment funds, resulting in a vicious cycle of inefficiency.
134 pages, 36 refs., 13 figs., 17 tabs., Language: Korean |
