Key indicators for the credit risk evaluation of clients and their changing characteristics

The credit risk of energy-using units (clients) is one of the main challenges in energy performance contracting projects. We develop a credit risk evaluation model for energy performance contracting projects that is optimized using rough set theory and random forest interpolation. The model is used to analyze 69 420 data entries from the Wind database and the Shanghai Stock Exchange for 178 listed companies (clients) with high energy consumption between 2007 and 2019. Our results show that the long-term capital debt ratio, current ratio, net profit growth rate, payable turnover ratio, asset–liability ratio, receivable turnover ratio, degree of operating leverage, cash ratio, operating profit margin, net sales margin and degree of financial leverage are key indicators closely related to the credit risk evaluation of clients. Debt-paying ability is the optimal primary indicator for the credit risk evaluation of clients, while the long-term capital debt ratio is the optimal key indicator. These key indicators have a low data dispersion and a relatively stable variation. Our model findings suggest ways to reduce the opportunistic behavior of clients and thus reduce the transaction costs of energy performance contracting projects, which could help to increase the motivation and confidence of project participants and improve the success rate of these projects.