News on the Corporate Social Credit System

by Felix Engelhardt

What do we already know?

The fact that the People's Republic of China has had a globally unique system for regulating the market and all the players operating in it for many years, and that this system is being continuously developed, is not in itself a novelty. Also, the main components (such as regulatory exchange of information, credit ratings and scores, blacklists and redlists, etc.), the functioning and the practical impact of this social credit system on companies ("Corporate Social Credit System", often abbreviated as "SCS" or "CSCS") have already been the subject of comprehensive analyses and assessments, including here, here, here and here. Therefore, we would like to refer to those articles for a more in-depth discussion of the CSCS.

With the announcement of the Opinions on Promoting Enterprise Credit Risk Management to Further Enhance Regulatory Effectiveness ("Opinions") on 13 January 2022, China's top market regulator, the State Administration for Market Regulation ("SAMR"), has now provided new insights into the concrete design of an important element of the CSCS.

What is new?

The Opinions are essentially an administrative document that is initially addressed only to all SAMR departments and local AMRs, i.e. they do not directly create rights or obligations vis-à-vis companies and individuals. Despite the fact that the Opinions are not binding outside the Chinese administration, they are of great importance for understanding the CSCS and for aligning business activities with it.

The Opinions describe and specify an internal administrative system for classifying all companies operating on the Chinese market into four risk classes (A, B, C, D). This is intended to make market monitoring more efficient, more accurate and faster, and to concentrate official resources in the right places. This appears reasonable in view of the approximately 150 million companies registered in China and the extent of potential (and actual) damage caused by illegal practices. Depending on the respective risk class, affected companies must expect either more frequent or less frequent inspections by the authorities.

In detail, the breakdown of risk classes is as follows:

• Risk class A (low risk) - number and frequency of random inspections reduced, except for specific complaints, notifications from other authorities, probles in Big Data monitoring;
• Risk class B (general risk) - normal frequency of random inspections;
• Risk class C (relatively high risk) - increased monitoring and frequency of random inspections;
• Risk class D (high risk) - strict and targeted monitoring and high number and frequency of inspections (including on-site).

The respective classification will be dynamic, at best in real time, scientifically sound, and uniform throughout the country. The latter aspect in particular is emphasized in the Opinions, since in the past there have often been significant differences in the selection of relevant criteria and the subsequent evaluation in different provinces or regions of China. A certain degree of uniformity is to be achieved by means of an index system at the central government level, to which local authorities are to be oriented on the one hand, but on the other hand are also to be given a certain amount of leeway to take local peculiarities into account.

The data on which the classification is based will be drawn from a variety of areas and sources so that the credit risk of a company can be comprehensively mapped. The sources of the data include information registered by companies themselves, administrative permits and licenses, information on intellectual property rights, administrative and court proceedings and sanctions, government lists (blacklists and redlists), the results of official controls and inspections, etc. The data used for the classification will also include sector-specific data, depending on the individual case. In addition to these general data, sector-specific data will also be included in the classification on a case-by-case basis, such as from the areas of food and pharmaceuticals, medical devices, general product quality or competition and anti-monopoly law. The Opinions emphasize that general risk classification does not replace sector-specific monitoring, but rather that these should be exercised alongside each other.

In line with China's overarching economic policy, the new classification system will also use future technologies, especially from the fields of big data, machine learning and artificial intelligence. In particular, classification is to be carried out automatically by algorithms, with the subsequent measures apparently (still) being selected and implemented exclusively by government officials.

When will the new classification system come into effect?

As the Opinions are not a legally binding document with external effect, they do not contain a specific date for their entry into force. In fact, the system has already been piloted in Shandong Province since 2019, and findings from this local pilot program have now been partially incorporated into the new system. However, the Opinions mention three milestones:

• By the end of 2022, a general enterprise risk classification system is to be established at the local level by the relevant authorities, all locally registered enterprises are to be classified according to the system, and other necessary preparations are to be made to implement the system;
• By the end of 2023, a functional enterprise risk classification system is to be established at the local level, which will be exercised jointly with sector-specific administration;
• Within the next 3 years, the enterprise risk classification system is to be fully integrated into government market regulation to identify risks early and manage them effectively.

This ambitious timetable shows that the classification system will very soon be an integral part of regulatory compliance in China.

What is our assessment?

First of all, we can say that the new classification system applies indiscriminately to purely Chinese as well as foreign-invested enterprises. This was already true for the application of the CSCS in other areas. Another positive aspect is that the Opinions contain an explicit commitment to grant companies sufficient time in their efforts to achieve compliance with the law, and even to actively support them in doing so. This makes sense, since self-regulation by the companies concerned can effectively contribute to the prevention of risks and the conservation of state resources.

However, it is problematic that the classification results are only shared internally within the authorities, which means that, according to our current state of knowledge, companies can only learn about their respective classification indirectly through increased controls. It is currently impossible to say whether this will remain the case in the future or whether companies will be given more opportunities to learn about their classification and thus proactively influence it. Another extremely critical aspect is the widespread use of AI algorithms to classify companies. In addition to opportunities for greater efficiency in public administration, the possibility of inaccurate classification of the companies concerned is very real. It remains to be seen how the state will ensure that the data fed in is error-free and complete, that the algorithms produce accurate results and that sufficient mechanisms are available to correct such errors. Finally, it remains to be hoped that the harmonization of market surveillance envisaged by the state can actually be realized and that there will not continue to be divergent or even contradictory decisions and official practices in different provinces and regions of China.

What can companies actively do?

Even though it appears that company classification will in future mainly be a government-internal process, companies can still actively prepare for and adapt to the new system. To this end, we recommend first of all (if not already done) to familiarize themselves with the basic principles and functioning of the CSCS. Only through a solid understanding of this far-flung administrative mechanism is it possible for companies to recognize the scope of the system and identify opportunities to influence the outcome. Next, identify areas in which the respective company has fallen short of legal requirements in the past (such as taxes, permits, product safety, occupational health and safety, etc.). The next step is to pay increased attention to the areas identified as problematic and to take prompt action to correct them. On the other side of the coin, certain activities and conditions can have a positive impact on the company's classification. For example, a company may benefit if grievances initially identified only internally are actively addressed or reported to the authorities on their own initiative. Finally, a comprehensive compliance system should be set up in the company - if one does not already exist - to identify risks from various areas of the company at an early stage and eliminate them effectively.