Summary:
Under the vision of carbon peaking and carbon neutrality, industrial enterprises are increasingly subject to carbon emission reduction constraints. Therefore, the credit rating that incorporates carbon element-related risks aims to reasonably distinguish the credit risk levels of different enterprises, and provides a reference for commercial banks to effectively warn enterprises of low-carbon transition risks and make green credit decisions. Based on the perspective of financial stability, this paper proposes a set of credit rating methods for carbon emission reduction performance of industrial enterprises, that is, while evaluating the performance of enterprises' carbon emission reduction, taking into account the solvency of corporate funds, and achieving a balance between ecological and economic benefits. The credit rating is divided into three steps, including the calculation of the credit levels of industrial enterprises, the credit rating of systemically important industrial enterprises, and the credit rating of all industrial enterprises. The above three parts are progressive, and constitutes the three rating steps. The first step is the calculation of credit levels for industrial enterprises. In this paper, the five financial indicators in the core credit indicator combination and the three low-carbon indicators, which are corporate carbon emission reduction potential, corporate carbon emission reduction capability and corporate environmental information disclosure level, are substituted into Structural Equation Modeling (SEM) to calculate the credit levels for industrial enterprises. The credit levels are the basis for credit rating, that is, the higher the credit level, the higher the credit rating; and vice versa. The second step is the credit rating of systemically important industrial enterprises. This step intends to construct a programming model (with the objective function of “the highest degree of matching between the credit rating of an enterprise and its anti-risk capability”, and to block each inter-enterprise credit risk contagion path as a constraint) to solve the optimal credit rating of enterprises. Therefore, all industrial enterprises are divided into different credit grade intervals to reduce the difficulty of classification, and the financial stability objective is integrated into the rating model from the perspective of credit risk contagion. The third step is the credit rating of all industrial enterprises. This step divides the remaining industrial enterprises into different credit grade intervals based on the credit levels of the enterprises, thereby simplifying the nine-class division problem into a binary division problem; at the same time, based on the “small range traversal +Sequential Forward Selection (SFS)”, the optimal critical samples between grades are searched to achieve credit ratings for all industrial enterprises. The main innovations are as follows. First, during the evaluation process, not only the vertical comparison of the corporate credit levels, but also the contagious effects of credit risk among enterprises, between enterprises and the banking system are considered to prevent the occurrence of systemic risks, thus incorporating financial stability objectives into the rating model. Specifically, as far as the relationship between enterprises and the banking system is concerned, this paper is based on the principle of “the higher credit rating of an enterprise, the stronger its ability to withstand the risk impact of the banking system”. The degree of matching between credit rating and the anti-risk ability is measured by the “inconsistency”, which is the objective function of the programming model. As far as the inter-enterprise association is concerned, based on the network analysis method, this paper constructs an association network among systemically important industrial enterprises, and then clarifies the contagion paths of credit risk among enterprises. On this basis, this paper blocks the paths for preventing systemic risks, which on the way of “dividing enterprisei and jinto different credit ratings”. And each path is a constraint of the programming model. Second, dividing all industrial enterprises into four credit grade intervals based on the credit ratings of systemically important industrial enterprises, and then convert the nine-class division problem into a binary division problem to avoid the “combinatorial explosion”. The division of credit rating of industrial enterprises can be regarded as an unconstrained combinatorial optimization problem. For the solution of this problem, the traversal method is the most reliable, that is, on the basis of considering all possible combinations of credit ratings of enterprises, the optimal solution that makes the evaluation function perform the best is obtained. However, the traversal method is limited to the case where the number of enterprisesn is small, and whenn is large, the method will lead to “combinatorial explosion” in the solution process of the optimization problem. Through the determination of the four credit rating interval structures, the problem of “dividing all industrial enterprises into 9 grades” is transformed into a binary division problem for the samples which come from the credit grade interval. This greatly reduces the difficulty of classification, and cleverly avoids the “ combinatorial explosion ” in the process of credit rating for all industrial enterprises. Third, searching for the optimal critical samples among different credit grades within the grade interval for binary division based on the “ small range traversal +Sequential Forward Selection”, so as to realize the credit ratings for all industrial enterprises. This idea not only avoids the risk of corporate default losses caused by inflated credit ratings to a certain extent, but also does not hinder the green and low-carbon transformation of industrial enterprises due to excessively stringent rating requirements.
邢秉昆. 碳减排约束下中国工业企业信用评级[J]. 金融研究, 2022, 509(11): 77-97.
XING Bingkun. Credit Rating of Chinese Industrial Enterprises under the Constraints of Carbon Emission Reduction. Journal of Financial Research, 2022, 509(11): 77-97.
Abdou, H.A. 2009, “Genetic Programming for Credit Scoring: The Case of Egyptian Public Sector Banks”, Expert Systems with Applications, 36(9): 11402～11417.
[13]
Attig, N., S.E. Ghoul, O. Guedhami, and J. Suh. 2013, “Corporate Social Responsibility and Credit Ratings”, Journal of Business Ethics, 117(4):679～694.
[14]
Baghai, R.P. and B. Becker. 2020, “Reputations and Credit Ratings: Evidence from Ccommercial Mortgage-backed Securities”, Journal of Financial Economics, 135(2):425～444.
[15]
Billio, M., M. Getmansky, A.W. Lo, and L. Pelizzon. 2012, “Econometric Measures of Connectedness and Systemic Risk in The Finance and Insurance Sectors” , Journal of Financial Economics, 104(3):535～559.
[16]
Guangming, G., X. Si, and G. Xun. 2018, “On The Value of Corporate Social Responsibility Disclosure: An Empirical Investigation of Corporate Bond Issues in China”, Journal of Business Ethics, 1(150):227～258.
[17]
Jing, J.B., W.W. Yan and X.M. Deng. 2021, “A Hybrid Model to Estimate Corporate Default Probabilities in China Based on Zero-price Probability Model and Long Short-term Memory”, Applied Economics Letters, 28(5) :413～420.
[18]
Liu, L., D. Luo, and L. Han, 2019, “Default Risk, State Ownership and the Cross-section of Stock Returns: Evidence from China”, Review of Quantitative Finance and Accounting, 53(4):933～966.
[19]
Zeidan, R.,C. Boechat,and A. Fleury. 2015, “Developing a Sustainability Credit Score System”,Journal of Business Ethics,127(2): 283～296.
