,

1. Bangor College, Central South University of Forestry and Technology, Changsha 410004, China; 2. Hunan Provincial Engineering Laboratory of Risk Management and Control Techniques for Innovative Enterprises, Changsha 410004, China; 3. College of Economics, Central South University of Forestry and Technology, Changsha 410004, China

Abstract　As leading enterprises supporting the agricultural development, agricultural listed companies play a great role, thus the study on their operating efficiency will have a strong demonstration effect. Using the data envelopment analysis (DEA) model, the operating efficiency of 20 agricultural listed companies was compared and analyzed. The results indicated that the overall efficiency of the agricultural listed companies is low. In agricultural listed companies with low or moderate overall efficiency, some agricultural companies are affected by the pure technical efficiency or scale efficiency, while other companies are affected by both factors, and accompanied with the increasing or decreasing efficiency along with the scale. Through in-depth analysis, it concluded that nine companies affected by pure technical efficiency have different degree of input redundancy or output deficiency. On this basis, it came up with pertinent recommendations for improvement.

Key words　DEA model, Agricultural listed companies, Operating efficiency

1　Introduction

Agricultural development concerns the increase of farmers’ income and rural development. At present, improving the quality and efficiency of agricultural supply is an important approach for China’s agricultural development. Agricultural listed companies play an important role in organizing farmer households’ production, raising the level of farmers’ organization, and technological innovation. With the gradual expansion of the asset size of agricultural listed companies, business activities have become increasingly complex, the sales network has expanded accordingly, and customer demand has increased, and agricultural listed companies have developed rapidly. However, at the same time, there are many problems in the process of development, such as the lack of ability of the corporate entities to evade market risks, operating losses, large market competition and low technological innovation capabilities, and low management level. Therefore, evaluation and study on the operating efficiency of agricultural listed companies are of great significance to transforming the economic growth mode, realizing the optimization and upgrading of industrial structure, and promoting the development of agricultural industrialization.

In foreign countries, the research on the operating efficiency started earlier. In the research of performance of agricultural enterprises, the DEA method was widely applied, but mainly focused on technical efficiency. Stephen Haaga[1]used the DEA model to evaluate the relative technical efficiency of agricultural production units. Since then, relevant scholars have used family farms and agricultural enterprises as evaluation objects to evaluate the technical efficiency of agricultural production in different regions[2-3]. Based on DEA model and through the qualitative and quantitative analysis, Veronika Fenyvesetal.[4]applied the rough set theory to analyze and evaluate China’s agricultural product network marketing performance.

Domestic research on the performance of agricultural listed companies mainly focuses on the performance evaluation methods of agricultural enterprises, the construction of agricultural performance evaluation systems, and the factors affecting the performance of agricultural companies. Selecting 13 financial indicators and applying principal component analysis (PCA), He Huiting and Liu Jian[5]empirically studied the economic performance of listed companies[5]. Taking China’s agricultural listed companies as the research object and using efficiency investment model or constructing the frontier function of production, Ding Zhu[6]analyzed causes and factors influencing efficiency of companies and came up with related recommendations accordingly. Combining economic performance, social performance and ecological performance evaluation, Zhang Meicheng[7]built a comprehensive model for the performance evaluation of agricultural listed companies and a corresponding indicator evaluation system, introduced the methods of economic value added, analytic hierarchy process, and related mathematical statistics for comprehensive performance evaluation of agricultural listed companies, and came up with corresponding promotion strategies and recommendations based on the analysis[7]. Taking agricultural companies listed in Shenzhen and Shanghai stock markets in 2009-2011, Wang Hongetal.[8]made a regression analysis on the relationship between management incentives and company performance, to prove that the management incentives could promote the improvement of performance of agricultural listed companies.

Based on the sample data and using the rotated component matrix, Huang Xiaoboetal.[9]obtained the variance contribution rate of six common factors, established comprehensive performance evaluation indicators for agricultural listed companies, so as to put forward recommendations for enterprises improving the operating performance. On the basis of the traditional operating performance evaluation indicator net profit after tax, Lei Na and Deng Shuhong[10]introduced EVA indicator and selected data of Shanghai and Shenzhen stock markets in 2010-2014, and made an evaluation of the performance of China’s agricultural listed companies. Using principal component analysis, selecting the data of 37 agricultural listed companies in 2012-2014, Wang Lei and Liu Huiping[11]evaluated their performance from profitability, debt repayment, operation, and development ability, and found the reasons for different levels of performance and came up with pertinent recommendations. Taking China’s agricultural listed companies, Chen Chengetal.[12]analyzed the impact of agricultural R&D investment on corporate performance and the lag effect, and put forward corresponding countermeasures. From the aspects of asset safety, operating efficiency, operating effectiveness and growth, Ai Xue[13]established performance evaluation indicators for agricultural listed companies, analyzed company performance and came up with recommendations. Through studying the effect of analysis diversification of data of agricultural listed companies on the corporate performance, Zhong Lihan and Yu Jing[14]found that diversification has a negative impact on corporate performance, and they put forward recommendations for improvement of diversified business strategies of China’s agricultural listed companies.

According to the above literature review, the performance evaluation methods of agricultural listed companies have their own advantages and disadvantages. The development of agricultural listed companies is influenced by different factors, and their production and operating efficiency changes and develop. Therefore, using the DEA model, we analyzed the development of 20 listed agricultural companies in China, evaluated their operating efficiency, to provide supplement and improvement for previous studies, and set foot on the present, and look forward to the future development.

2　Model and indicators

2.1ModelbuildingThrough the CCR model, the operating efficiency value of the selected agricultural listed company can be judged by its comprehensive efficiency, and the scale of return of each agricultural listed company can be obtained. The judgment rules are as follows: (i) whena*=1,sio-=0,sio+=0, it can be determined that the comprehensive efficiency of the agricultural listed company is effective for DEA, in other words, the agricultural listed company has economies of scale and the technology is effective; (ii) whenα*= 1 and at least one ofsio-andsio+is not 0, it can be judged that the agricultural listed company does not have economies of scale or technical inefficiencies, resulting in no comprehensive efficiency; (iii) whenα*<1, it can be deemed that the comprehensive efficiency of the agricultural listed company is ineffective for DEA.

2.1.1Building of DEA model. DEA (Data Envelopment Analysis) is a new field of cross-disciplinary research in Operations Research, Management, and Mathematical Economics. It is a new efficiency evaluation method developed by the well-known operational researcher Charnes and Cooperetal. based on the concept of "relative efficiency". DEA is to evaluate the relative effectiveness (DEA validity) if decision making units (DMUs) of many inputs and outputs through a mathematical programming model. According to the data observed by each DMU, it is able to determine whether the DMU is valid for DEA. Essentially, it is to determine whether the DMU is located on the "frontier" of possible production set. When using DEA to evaluate the efficiency of DMU, it is able to obtain much management information with profound economic implications and background in economics. The DEA evaluation model is particularly suitable for complex systems with multiple input variables and output variables. It evaluates both the scale effectiveness and the technical effectiveness of DMUs[15].

The application field of DEA method is continuously expanding, and it has been applied to the evaluation of policy effects, evaluation of business efficiency, and evaluation of industrial development benefits, while the development of DEAP software and EMS software has made the application of DEA model more convenient and accurate. From the perspective of the operating efficiency of agricultural listed companies, the basis is the evaluation of the operating efficiency of the company. Therefore, the DEA model was selected for research. Generally, the DEA method is used to evaluate the operating efficiency of a company, and it is mainly divided into two evaluation models, namely, the CCR model and the BCC model.

2.1.2Building of the CCR model. For evaluating the operating efficiency of agricultural listed companies using the CCR model, the calculation principles and internal logic steps are as follows: assume that there are m listed agricultural companies to be analyzed, namely, m DMUs to be determined for operating efficiency, denote these companies as DMUk, and 1 ≤k≤m. At the same time, assuming that each agricultural listed company hasxkinds of inputs andykinds of outputs, the input vector of thek-th agricultural listed company can be expressed asRk= (r1k,r2k, ....rxk)T,k= 1,2,…,m, and that the output vector of thek-th agricultural listed company can be expressed byCk= (c1k,c2k, ...cxk)T,k= 1,2,...,m. The corresponding weight is denoted asq= (q1,q2,...,q).

Taking the operating efficiency index of selected several agricultural listed companies, taking the operating efficiency index of the ko-th agricultural listed company as the target, through introducing the dual programming model of the corresponding evaluation objects, we obtained the slack variables+and surplus variables-to judge the room for improvement, and obtained the following equality constraints:

(1)

where min represents the minimum value;s.t. represents the constraint condition; α*represents the overall efficiency evaluation result;εrepresents the non-Archimedean infinitesimal quantity;RkandCkdenote the input and output vectors respectively;s+ands-denote the introduced slack variable and the surplus variable;λkdenotes the combination coefficient of each evaluation unit[13].

2.1.3Building of the BCC model. The BCC model is based on an in-depth or further analysis of evaluation results of the CCR model. Using the CCR model, it is able to determine whether the overall operating efficiency of a specific listed company is DEA valid, and the BCC model can conduct in-depth analysis of non-DEA effective agricultural listed companies calculated by CCR, and can clearly know what causes the company’s DEA to be invalid.

The mathematical expression of the BCC model is as follows[25]:

(2)

whereβ*denotes the evaluation results of pure technical efficiency, and the meanings of min,s.t.,ε,Rk,Ck,s+,s-, andλkare the same as in CCR model.

2.2IndicatorselectionWhen using DEA for efficiency analysis, ensure that the number of analysis objects is more than twice the input and output indicators, because there are 20 selected research objects (DMUs). Therefore, four input and output indicators are separately selected to ensure the rational and scientific selection of indicators. The specific indicator building is shown in Table 1.

Table 1　Building of indicators for DMUs

The DEA model itself is highly sensitive to input and output, so we applied the DEA method to evaluate the operating efficiency of listed companies. Besides, the evaluation indicator system is the key to the effective use of the method. Different evaluation indicator systems will bring different evaluation results. Therefore, based on the literature review, we obtained implication and references, and selected scientific indicators and built an effective indicator system. In building the indicator system for evaluating the operating efficiency of agricultural listed companies, two key points are considered. (i) Applicability of indicators. In general, when applying DEA for efficiency analysis, it is necessary to ensure that the number of analysis objects (DMUs) is more than twice the input and output indicators, because there are 20 selected research objects (DMUs). In this study, we selected four input and output indicators to ensure the rational and scientific selection of indicators. (ii) Availability of data. Fully considering the applicability of indicators and the validity of the annual report data of agricultural listed companies, we selected input indicators including total assets, fixed assets, current assets, and operation cost. The first two indicators reflect the company’s overall strength and are the basis for stable economic benefits, and the latter two indicators reflect the continuous investment of listed companies. For output indicators, we selected the operation income, operation profit, net profit, and return on net assets. The first two indicators are considered from the market efficiency, and the latter two are selected from profitability.

3　Indicator and data source

3.1DatasourceThe data selected in this study came from the data issued by official websites of listed companies and financial data of cninfo (http://www.cninfo.com.cn), and agricultural listed companies we studied are mainly divided into traditional agricultural companies, seed companies, and breeding companies. Traditional agricultural companies mainly include Beidahuang, Xinjiang Tarim Agricultural Comprehensive Development, Xinjiang Korla Pear, and Yasheng Industrial; seed companies include Fengle Seed, Longping High-Tech, and Denghai Seed; breeding companies include New Wellful, Fucheng Wufeng Food, Huaying Agriculture, Sunner Development, Minhe Animal Husbandry, and Haikou Agriculture & Industry & Trade (Luoniushan). Finally, we selected following 20 agricultural listed companies to conduct empirical analysis: New Wellful, Muyuan Food, Truein Group, Sunner Development, Longping High-Tech, Minhe Animal Husbandry, Beidahuang, Yisheng Livestock & Poultry Breeding, Denghai Seed, Fucheng Wufeng Food, Xiantan, Huaying Agriculture, Yasheng Group, Wanxiang Denong, Yunnan Investment Ecology and Environment Technology, Luoniushan, Xinjiang Korla Pear, Fengle Seed, Xinjiang Western Animal Husbandry, and Win-all High-tech Seed.

3.2DataprocessingBased on the selected input and output indicator data of the agricultural listed companies in 2016, the data were brought into the model and DEAP 2.1 software was used to calculate the overall efficiencyα*of the 20 agricultural listed companies in 2016 and the pure technical efficiencyβ*and pure scale efficiencyλ*, the results were shown in Table 2.

Table 2　Calculation results of DEA efficiency of agricultural listed companies

4　Empirical results and analysis

4.1ExperimentalresultsWith the aid of DEA2.1 software, we calculated the DEA effectiveness of 20 agricultural listed companies in 2016 based on the CCR and BCC models. The results are shown in Table 2. From Table 2, it can be seen that Beidahuang, Yasheng Group, Wanxiang Denong, Fengle Seed, and Xinjiang Western Animal Husbandry realized DEA effective in both the technical efficiency and scale efficiency. In the rest 15 companies, four companies (Longping High-Tech, Yisheng Livestock & Poultry Breeding, Denghai Seed, and Luoniushan) were effective in technology but not in scale, so the overall efficiency was not effective, because these companies were influenced by natural factors such as crop rotation and harvest date; six companies (Muyuan Food, Truein Group, Huaying Agriculture, Yunnan Investment Ecology and Environment Technology, and Xinjiang Korla Pear) were effective in scale but not in technology, so the overall efficiency was not effective, these companies had certain scale in respective field, but had no advantage in the technology. Five companies were not effective in the overall efficiency because of ineffective in both the technology and the scale, namely, New Wellful, Minhe Animal Husbandry, Fucheng Wufeng Food, Xiantan, Win-all High-tech Seed. These companies are restricted by the production scale due to natural laws, and their technologies are mature, it is difficult to form unique technical efficiency, so the technical is not high.

Based on the BCC model, we calculated the values of surplus variables-and slack variables+, as shown in Table 3. According to Table 3, it is necessary to make adjustment for nine listed companies whose technologies were ineffective. Nine companies had different levels of input redundancy or insufficient output. Companies having input redundancy are mainly Minhe Animal Husbandry, Yisheng Livestock & Poultry Breeding, and Win-all High-tech Seed. The reason is mainly that the first two companies belong to the leading enterprises of agricultural industrialization. With the increase in the sales both at home and abroad, the input also increases. However, valuing the output but belittling the quality, agricultural listed companies have problems of frequent change in raising funds, deviating from the developing objective of promoting agricultural industrialized operation, and correlation between market reaction and performance impact. Therefore, from the perspective of controlling investment, promoting product transformation and upgrading, creating a product brand and expanding the perspective of overseas markets, it is required to alleviate the problem of low operating efficiency caused by input redundancy. Companies with insufficient outputs mainly include New Wellful, Minhe Animal Husbandry, Yisheng Livestock & Poultry Breeding, Fucheng Wufeng Food, Xiantan, and Win-all High-tech Seed. This is because with the development of global economic integration, under the influence of the wave of cross-border mergers and acquisitions, China’s agricultural listed companies have been squeezed by foreign capital and products. For example, mergers and acquisitions in developed countries, cross-group mergers and acquisitions, and the pressure of mergers and acquisitions of listed companies in the same industry. All these make the agricultural listed companies change the fund raising frequently; at the same time, the agricultural listed companies are also affected by the macroeconomic environment, the industry development environment and other factors, which directly and indirectly affect their operating performance. These companies have relatively single businesses, and most companies have specialized production. On the one hand, the industrial demand or price are low, and market capital is short; on the other hand, due to influence of long production cycle and large supply, the output efficiency was affected to a certain extent. In summary, it is necessary to optimize the input and output efficiency of nine companies in order to achieve effective DEA in both technical efficiency and scale efficiency.

Table 3　Values and distribution of slack variables for ineffective pure technical efficiency of agricultural listed companies

4.2AnalysisofoverallefficiencyThe overall efficiency values of listed agricultural companies are separately summarized and divided into four stages of efficiency. In these companies, five companies had overall efficiencyα= 1, namely, Beidahuang, Yasheng Group, Wanxiang Denong, Fengle Seed, and Xinjiang Western Animal Husbandry. The input-output ratio of these five companies is moderate, showing a strong input-output efficiency, and it is required to keep such a good development trend. Sunner Development, Longping High-tech, and Xinjiang Korla Pear belonged to mildly DEA ineffective (0.8 ≤α< 1). In these five companies, four companies had moderately DEA ineffective (0.5 ≤α<0.8), namely, Muyuan Food, Fucheng Wufeng Food, Huaying Agriculture, and Yunnan Investment Ecology and Environment Technology. The moderate DEA ineffectiveness of Muyuan Food and Fucheng Wufeng Food is inconsistent with the actual situation of their operating scale, mainly associated with the selected input and output efficiency indicators. Six companies had seriously DEA ineffective (α<0.5), including New Wellful, Truein Group, Minhe Animal Husbandry, Yisheng Livestock & Poultry Breeding, Xiantan, and Win-all High-tech Seed. This indicates that the overall efficiency of these companies’ inputs and outputs are low, and the allocation of resources needs to be further improved. Generally, most companies have low overall efficiency, and the overall development status is not optimistic because of the external factors such as the long-term operating cycle, weak anti-risk capability and other external factors of the agricultural listed companies, as well as the industrial environment and capital markets.

4.3AnalysisofpuretechnicalefficiencyWe summarized and divided the pure technical efficiency of agricultural listed companies into four stages. From the data, seven companies reached optimal technology (β= 1). These seven companies are Beidahuang, Denghai Seed, Yasheng Group, Wanxiang Denong, Luoniushan, Fengle Seed, and Xinjiang Western Animal Husbandry. Eight companies had mildly ineffective pure technical efficiency (0.8 ≤β<1), namely, New Wellful, Sunner Development, Longping High-tech, Minhe Animal Husbandry, Fucheng Wufeng Food, Xiantan, Xinjiang Korla Pear, and Win-all High-tech Seed. Three companies (Muyuan Food, Huaying Agriculture, and Yunnan Investment Ecology and Environment Technology) had moderately ineffective pure technical efficiency in the range of[0.5, 0.8). Two companies (Truein Group and Yisheng Livestock & Poultry Breeding) had seriously ineffective pure technical efficiency (β< 0.5). The technical efficiency of most companies is high, indicating that the overall technical level and management level of agricultural listed companies are relatively high.

4.4AnalysisofreturnstoscaleBased on the DEA model, the types of scale efficiency can be divided into three stages: the scale efficiency is constant, increasing, and decreasing. The stage of scale efficiency is the stage when the input and output of an enterprise reaches an optimal state. In the process of decreasing scale efficiency, increasing the input will not only lead to an increase in corporate profits, but will reduce the company’s income. In this regard, from the perspective of returns to scale, it is possible to provide policy recommendations for increasing or decreasing input. Through integration of DEA efficiency calculation of agricultural listed companies, it is able to obtain the distribution of changes in returns to scale of agricultural listed companies. The scale efficiency of Beidahuang, Denghai Seed, Yasheng Group, Wanxiang Denong, Luoniushan, Fengle Seed, and Xinjiang Western Animal Husbandry was constant, accounting for 35% of the total, their input and output realized excellent matching, the efficiency was high, and in a short term, it is recommended to keep such excellent resource allocation state. New Wellful, Fucheng Wufeng, Xinjiang Korla Pear, and Win-all High-tech Seed had increasing returns to scale, accounting for 20% of the total. Muyuan Food, Truein Group, Sunner Development, Longping High-tech, Minhe Animal Husbandry, Yisheng Livestock & Poultry Breeding, Huaying Agriculture, and Yunnan Investment Ecology and Environment Technology had decreasing returns to scale, accounting for 45% of the total, possibly because of low efficiency of diversified businesses and downward trend of linked industries. These nine companies should optimize resource allocation, especially the fund allocation to realize increase of operating efficiency. In general, agricultural listed companies should combine their own conditions with the development of the industry to launch a modest scale, reduce projects with excessive capacity, and make proper resource allocation.

5　Conclusions

The performance evaluation of agricultural enterprises is an important subject worthy of study, and has great research value and significance. With reference to the experience of corporate performance evaluation, using the DEA model as a research method, we made an in-depth discussion of operating efficiency of 20 agricultural listed companies in China, and built a performance evaluation model suitable for China’s agricultural listed companies. Through empirical analysis, it could be concluded that the DEA model is objective, applicable, and effective for evaluating the performance of agricultural enterprises. The overall operating efficiency of agricultural listed companies is generally low, and there is still much room for improvement in the operating efficiency of agricultural listed companies. Listed companies with ineffective DEA may be influenced by scale or technology, indicating that these companies should make further strategic considerations in the scale control or technical improvement.