Preliminary Research On The Identification And Separation Of Performance Noise

Government performance is the main criterion to evaluate the efficiency of government. Continuous improvement of government performance is the target of the public management reform. Many countries, including China, have gradually carried out the reform of the government performance management. But practice shows that the current performance management reform does not achieve the ideal effect. The performance evaluation is an important part of performance management. The ultimate goal of performance evaluation is to implement the performance improvement through the application of the performance information in the level of accountability and budget decisions. This paper attempts to analyze the potential problems in process of the performance information applied in the level of the accountability and budget decisions from the perspective of the noise and then find solutions. Under the background of the reform of financial performance evaluation and management deepened, how to properly handle the problem of noise has become a major issue in theory and practice.Accountability and budget decisions produce the performance incentive through the use of information from performance evaluation. if the performance evaluation get disturbed by the interference of noise, thus the performance incentive produced by accountability and budget decisions will also be distorted. The Third Plenary Session of the 18 th Central Committee of the Communist Party of China put forward to improve and implement the leading accountability. Good accountability requires to distinguish controllability and uncontrollability of the responsibility of the administrators, administrators should only take the accountability for controllable affairs within the scope of responsibility. Wise decision also requires to clearly identify the correlations of the input- output- result. The basic logic of performance budgeting is to allocate scarce resources according to the level of performance, if not clearly define or classify the source of the performance, making right resource allocation decisions will be impossible. Correlation is the basic principle of decision, if the output and the outcome largely depend on the effect of noise, thus this decision is not scientific. General speaking, the influence of the noise should not be listed as the scope of responsibility of administrators. The correlation of the noise and performance results should be considered when making budgeting decisions. Thus, whether building the accountability mechanism or making budget decisions requires to identify and separate the impact of noise on the result, and the controllability and the correlation have become the two important criteria to define whether it is subject efforts or noise.Unfortunately, although there are a lot of literature on performance evaluation and management, there is little research on noise problem. In the last ten years, the foreign research on performance evaluation focus on the integration of performance evaluation, namely comprehensively evaluate the performance. Foreign study is not only confined to the construction of evaluation index and the performance evaluation itself, but also make comprehensive consideration of subjective factors such as management ability, willingness, and external environment factors such as political and social influence on performance. But the existing literature doesn’t systematically define the meaning of the noise, and rarely take it into the consideration of performance evaluation. Thus, the effect and influence of noise in the performance management process, has not yet entered the domestic mainstream literature research field. In this paper, the research will help make up for the domestic research about the problem of noise in performance managementThis paper mainly consists of six parts. The first chapter elaborates the background and significance of this research, combines the related literature among domestic and abroad, and introduces the research ideas and methods. The second chapter put forward that the noise can make effect on the formation of performance, and also the result of the performance evaluation and the application of performance information, through the analysis of the four chain link of performance results(input-output- results- impact). Therefore, it is necessary to identify and separate the noise in the performance evaluation process. According to the principle of the controllability and correlation, the driving factors of performance are divided into the subjective efforts and the noise, and furthermore making more detailed classification.On this basis, the third chapter specially makes analysis of the influence on accountability and budget decision-making process caused by the noise. In terms of level of accountability, first of all, the existence of noise enlarges the scope of accountability of administrators, administrators should be responsible only for the subjective manageable behavior and results, and shouldn’t be responsible for the result of the uncontrolled factors. Secondly, the accountability achieves incentive power through the reward and constraint mechanism, and the interference of noise will affect the choice of the incentive method and the level of the incentive. In terms of budget decisions, performance information is one of the signals of budget decisions, it leads the budget allocation of scarce financial resources. First of all, the interference of noise can bring error performance evaluation results, misleading the allocation by the principle of rewarding good performance and punishing poor performance. Secondly, the noise distorts the rate of return on public spending, affecting the allocation of financial resources into more efficient fields, thus misleading the operational decisions. Only making good use of the information of the performance evaluation on the level of accountability and the budget decisions can achieve the goal of improving performance, and the noise interference can’t be overlooked in the process of accountability and budget decisions. Therefore, it is necessary to identify and separate the noise in the process of performance evaluation.Subsequently, based on data envelopment analysis(DEA) and the regression method in the fourth chapter, the preliminary model was established. In order to obtain more accurate performance information, the effects on government performance evaluation information of noise are identified and separated. One of the important problems in the process of noise identification and separation, is the quantified of noise interference. Noise involved in a wild field, quantifiable degree of all kinds of noise is also different. The noise in natural environment is more easily acquired and quantified. Therefore, the fifth chapter choose the air quality management in Beijing environmental protection department as a case study, hope to confirm and eliminate the influence of noise. Empirical results validated the proposed hypothesis that the noise had significant influence on the performance of government in air quality, so it is necessary to spun it off from the the performance evaluation.Finally, based on the separation of the noise, the sixth chapter proposed some proposals to improve the accountability and budget making. About accountability, first of all, performance accountability incentive requires to define the scope of accountability clearly, and to choose the right incentive method and degree; the second, constraint is an implementation way of accountability incentive, this article based on the viewpoint of strengthening fiscal transparency and citizen participation, to study the effect of constraints on how to strengthen accountability. The ultimate goal of performance evaluation is not the evaluation of performance, but to improve performance through the application of the evaluation results. About Budget decisions, it includes two basic types: the configuration decisions and the operating decisions. The former involves how to allocate scarce resources among different purposes, and the latter refers to how to get the ideal output and results at reasonable cost. Because of the noise interference misled two decisions, only based on the performance information that without the influence of noise, can performance evaluation provide reliable technical support to budget making.Only improved accountability and decision-making level, can the application of the performance evaluation information really achieve the goal of improving performance. Clearly, the accuracy and reliability of information is critical to the performance management. In order to obtain more accurate and more availability of performance information, it is necessary to identify and eliminate the impact of noise in the performance evaluation. In this paper, DEA model can calculate local government performance evaluation, then the paper use regression to analyze performance ratings and driving factors including all kinds of noises, and also distinguish and determine the effect of all kinds of performance factors. In addition, the idea which based on the separation of the noise, provides a new direction for the reform practice of performance management in our country. But it should be pointed out that, the model proposed in the article is a preliminary exploration research on effect of the noise in the process of performance process. The public sector is large and complicating, the separation model cannot meet the requirement in practice. In addition, based on the advantages of easily quantified of natural environment noise, this paper selects the Beijing air quality management case as an empirical study. But not limited to the natural environment noise, other kinds of noises also have an effect on the performance information. Therefore, according to the specific conditions of different departments, it is necessary to have a detailed research on the definition and classification of the noise. Computing performance ratings should also be discussed, and other methods to realize the separation of noise effects at the same time, to improve the model of the separation in performance evaluation.