Evolution of the Social Credit System: Three Functions through the Lens of Public Law

Essay for discussion at the Interdisciplinary Symposium “Super-Scoring? Data-driven Societal Technologies in China and Western-style Democracies as A New Challenge for Education.” (Cologne, 11 October 2019). The essay can be downloaded here as a PDF-File.

by Yongxi Chen

China’s Social Credit System (SCS) is widely regarded as a ground-breaking form of data-driven governance in media reports. The term “social credit system” as used in academic discourses also denotes a comprehensive networked system of behaviour rating and responsibility placing. The SCS in practice, however, is still taking shape, with official documents interpreting the concept of “social credit” differently and associating the system with changing policy goals. This essay seeks to provide a pathfinder in the labyrinth of policies and pilots concerning social credit in China by outlining three primary functions served by the SCS during its evolution. The analysis here does not intend to be exhaustive or technology-centric. It focuses on the legal nature of essential measures of assessment and punishment undertaken by different entities within the changing boundary of the system. The purpose is to capture better the challenges posed by the SCS to individuals’ rights and the limits of legal redress under China’s changing politico-legal landscape.

As Financial Credit Rating

A notion of social credit was initially introduced in the early 2000s to steer the economic reforms that increased the financial creditworthiness of businesses and individuals. [i] To construct a mechanism of “financial credit rating” (zhengxin in Chinese) that resembles its western counterpart was the primary goal of social credit projects conceived by national policy-makers for a decade. After 2011, directives of the Chinese Communist Party (CCP) and documents of the central government gradually perceived social credit as a tool for market regulation as well as social governance. The semantic expansion of social credit was formally endorsed in 2014 when the State Council promulgated the Planning Outline for Construction of the Social Credit System (”Outline 2014” hereafter).

It should be noted that financial creditworthiness assessment is a part of the current SCS and operates in parallel to other types of trustworthiness assessment (discussed in the following sections). From the perspective of the division of labour, the Central People’s Bank continues to be responsible for regulating credit rating businesses, while the National Reform and Development Commission (NRDC) is currently in charge of supervising other types of trustworthiness assessment. Policy coordination takes place at the Inter-ministerial Conference for SCS. In addition, credit scoring services have been developed by fintech companies resorting to big data analytics of users’ consumption and social behaviours, including the eye-catching Sesame Credit operated by the Chinese internet giant Alibaba. For the time being, such services are regarded as financial credit rating and subject to the licensing and supervision of the central bank.

The assessment of creditworthiness is governed by the Regulations on the Administration of Credit Investigation Industry 2013, which protects individuals’ rights concerning the collection and processing of their financial credit data. Insofar as the credit investigation institutions and entities providing credit data (e.g. commercial banks) are both private bodies, individuals enjoy civil rights concerning the protection of personal credit data. Such rights include, among others, consent must be sought of the use of one’s credit records, and individuals have a right to access and rectify those records.

Nevertheless, uncertainty surrounds the spillover effects of the new functions of the SCS on financial credit rating. The State Council’s guiding opinions on combined punishments for untrustworthy persons provide that the government should guide “the risk-based pricing in financial institutions so as to raise the loan rates and insurance premium rates in respect of untrustworthy entities, or impose restraints on loans [and] other [financial] services provided to them.” Under what circumstances a person’s non-financial “trustworthiness” should affect his or her financial creditworthiness is a profound and controversial policy issue which entails public inputs and careful deliberation. The rule-making process in this regard remains closed and opaque, which weakens accountability. Furthermore, it is difficult for individuals to challenge the credit rating results affected by these “coordinative arrangements” because the acts of administrative guidance are usually excluded from the scope of administrative litigation.

As Coercive Measures to Enforce Legal Rules

Purpose and Form

Towards the end of the 2010s, a non-economic dimension was introduced to the concept of social credit. In official and academic discourses, connotations of this concept shift from the narrow idea about “creditworthiness” to a broader albeit more ambiguous idea about “trustworthiness”, the latter encompassing the qualities of honesty (cheng) and keeping promises (xin) with respect to fulfilling contractual commitments and legal duties.

The Outline 2014 emphasizes the fulfilment of legal duties as the prominent quality of “trustworthiness” (shouxin, alternatively translated as “keeping faith [with others]” or “trust-keeping”). Therefore, imposing coercive measures to compel compliance with legal rules has become the core of all local pilots implementing the Outline. This function is vividly captured by the slogan appearing in most official documents: “Being untrustworthy on one matter, facing constraints everywhere.” (Yichu shixin, chuchu shouzu) In this policy context, “untrustworthiness” (shixin, attentively translated as “breaking faith”) is equated with infractions, in particular, the violations of the law which have caused long-standing economic and social issues. Natural and legal persons committing these infractions are labelled as “untrustworthy” and subject to the mechanism called “combined punishment” (lianhe chengjie, alternatively translated as “joint punishment” or “cooperative discipline”), under which coercive measures are concurrently imposed by different authorities to affect their interests across various domains, like a ripple effect.

The infractions that render a person untrustworthy are specified in a sector-specific manner, usually in the form of memorandums jointly issued by various authorities. By May 2019, 43 memorandums have been promulgated at the central level, covering sectors such as food safety (2016), wage arrears of migrant workers (2017) and intellectual properties protection (2018). Countless inter-departmental memorandums are also formulated at local levels.

Depending on the nature of the legal rules that are violated, combined punishments serve two separate purposes and reinforces different state powers. The first purpose is to overcome the chronic difficulty in executing effective judgements by the courts. The most conspicuous punishments for “judgement defaulters” are restrictions on luxury consumption and denial of access to plains and high-speed trains. It is noteworthy that travel bans have gradually extended to defaulters of liabilities to pay tax or administrative fees and thereby served the second purpose, that is to reverse the poor implementation of the law by administrative agencies and enforce liabilities determined by the government. Combined punishments for this purpose vary a lot and cover a broader spectrum of coercion. They include, to name a few, (1) restrictions affecting the untrustworthy person’s economic opportunities, such as establishing companies and raising finance; (2) denial of access to market sectors such as food, public utilities or banking; (3) disqualification for receiving government subsidies; (4) restrictions on participation in government procurement and transaction of state-owned properties; (5) rendering the untrustworthy persona illegible for honorary titles; (6) disclosure of the name of the person on major news portals and the national website of the SCS.

Roles of Data and Scoring

Combined punishments are preconditioned by cross-sector sharing of behavioural records. Unlike consequences of the financial credit rating, however, in most cases, combined punishments are triggered not by scoring but by blacklisting, i.e. the matching of persons and specified infractions. Blacklists concerning various sectors have proliferated since 2017. While credit scoring has been slowly rolling out in over a dozen cities, its consequences remain limited to offering rewards to people for taking socially beneficiary actions without direct impact on others’ rights. In contrast, blacklisting-triggered punishments have an enormous impact on individuals’ rights.

Impact

Combined punishments for judgement defaulters are widely considered as legitimate insofar as they derive from the judicial power to ensure execution of binding rulings and uphold the authority of law. There is a growing concern, however, about inflicting punishments on persons labelled as untrustworthy by administrative agencies. A substantial part of the punishments in this group are in nature administrative penalties or licensing whose conditions shall be provided by laws or regulations only. Creating such punishments through memorandums (which are not a source of law) contravenes the principle of the reservation of the law (similar to the German principle of Vorbehalt des Gesetzes). With regard to the punishments that do not belong to orthodox categories of administrative act and fall in the scope of agency discretion, they are plagued by the problems of irrelevancy and proportionality.

The infliction of punishment is based on the trustworthiness of a targeted infraction, different from the legal liability for the infraction. The trustworthiness seems to have ignored the differences in the nature and severity of the infractions. This de-contextualised approach affects the relevance and commensuration of punishment. First, relevance is sometimes difficult to see between an infraction in Sector A and the punishments ‘jointly’ inflicted by another authority regulating Sector B. For instance, under the Memorandum on Combined Punishments concerning Untrustworthy Persons in Marriage Registration, an individual who makes false statements about his or her consanguineous relation with the spouse should be barred from registering his or her enterprise as an “Authorised Economic Operator” with the customs authority. The relationship is obscure between a business entity’s conducts pertaining to the customs code and its legal representative’s breach of eugenic policies. Second, severe punishments may apply indiscriminately to infractions of different gravity. For example, “seriously untrustworthy” people excluded from high-speed trains encompass not only those who have refused to pay overdue debts as ordered by the courts, but also persons in charge of the social security service providers (such as hospitals) which have violated service agreements, and those who have been fined for smoking in rail-cars. It is questionable whether the three kinds of infractions share the same shade of intent or have similarly serious consequences.

Remedy

Although various rights are affected by the trustworthiness assessment in question, remedies are inadequate due to limitations of the legislation and judicial review.

Data privacy is poorly protected. Under the Chinese legal doctrine, information concerning violations of the law is not regarded as privacy. And in the absence of statutory principles of data protection, government agencies can disclose and use the infraction records without being confined to the individuals’ consent or the original purposes of data collection. Following the system of financial credit rating, local SCS legislation entitles “untrustworthy” persons to access their infraction records and request for correcting the inaccurate ones, which are essentially data subject’s rights. Nevertheless, such rights address merely the issue of factual accuracy in the infraction records. The legality of the findings of infraction – which involve subjective judgements of administrative agencies – cannot be challenged through data correction procedures.

Reputations of untrustworthy persons also suffer from the publication of their names on blacklists or other media, sometimes even without their knowledge. The courts have held that public naming and shaming do not amount to an administrative penalty and tended not to entitle the persons concerned to the rights to be notified and to hearing in respect of their reputation interests.[ii]

With regards to combined punishments that adversely affect other substantive rights, the courts have preferred formal review to robust substantive scrutiny. Coercive measures were quashed on procedural grounds[iii] or because of the agency’s failure to cite the legal basis[iv]. However, combined punishments based entirely on policy documents of local governments without legal authorisation were upheld[v], with the principle of reservation of the law ignored. Important grounds of judicial review, such as those of relevance and proportionality, have yet to be entertained in any reported case.

As Instruments to Strengthen Ideological Governance

Purpose and Form

The experiments on the SCS coincide with the adjustment of governance strategies under the CCP’s new leadership. An intention to employ the system for underpinning ideological control of the society emerged under the Outline 2014 and gained policy momentum in 2016.

The Party’s Central Committee and the State Council jointly issued in 2016 the Opinions on Integrating Core Socialist Values into the Building of the Rule of Law, which sets the SCS as an institutional device for restricting and punishing behaviours that contravene core socialist values. The values embody the official ideology as confirmed by the CCP 18th Congress and contain 12 major elements, including for example integrity (chengxin), professional dedication, friendliness and patriotism. The Party commands that such values be embraced by legislation and public policies and adhered to in social governance. Against this background, the central government has made special arrangements under the SCS to evaluate the compliance with ethics in selected fields and, in this policy context, has increasingly used the terms of “integrity system” and “personal integrity”. For analytical purposes, such arrangements are called “integrity assessment” hereafter.

The key fields identified in different national policy documents include: (1) immoral behaviours that have bred public discontent, such as charity frauds and falsifications in marriage registration; (2) conducts of professional groups which offer market-oriented services, such as lawyers, doctors, accountants, financial agents, etc.; (3) activities in the education sector, especially those of teachers and students; (4) business practices of enterprises that fall in the purview of self-regulation of industrial associations.

In contrast to the trustworthiness assessment for enforcing legal rules, “integrity assessment” involves more stakeholders in defining the assessment criteria and taking follow-up actions. First, the assessment criteria, i.e. ethics, derive not from enacted legal rules but positive behavioural standards set otherwise by public authorities. It is noteworthy that Party agencies and “people’s organisations” (which operate under the direct supervision of the Party) have actively participated in making the guiding principles for the criteria together with other state organs, which conveniently brings in the Party’s explanations of the ethics concerned in the light of socialist values. Second, results of “integrity assessment” will lead to combined punishments and rewards provided by bodies in both the public and private sector, in particular the self-regulatory organisations of professionals and business entities. While these private organisations probably enjoy latitude in providing incentives to their “trustworthy” members, it is unclear how much discretion they are allowed with regard to undertaking disciplinary or restrictive measures against the members rated as “seriously untrustworthy” according to the state-imposed standards of integrity.

Roles of Data and Scoring

“Integrity assessment” depends not so much on the cross-sector sharing of digitized records. Instead, it tresses the intensive profiling of persons in a given sector with regard to the targeted behaviours. The central government explicitly requires individualized “integrity dossiers” (chengxin dang’an) to be complied about students, “credit records” about the enumerated professionals, and “credit dossiers” about members of industrial associations. Profiling as such will make scoring easier. Since the assessment arrangements are at an early stage, the role of algorithms remains to be observed. Theoretically, the results of various types of “integrity assessment” are to be integrated into the unitary SCS platform. Depending on the policy orientation, in the future, they might be correlated with the results of trustworthiness assessment dedicated to enforcing legal rules and thereby regarded as additional grounds of punishments to be imposed by administrative or judicial authorities.

Impact and Remedy

The majority of the behaviours subject to “integrity assessment” are not regulated by legal rules, which implies that the modernized legal system refrains from interfering with the social relations concerned and allows self-governance in relevant professional or industrial communities. By prescribing positive standards for such behaviours, the SCS risks marginalizing or even substituting the behavioural standards that are agreed on or spontaneously formed within those communities (e.g. moral principles, professional ethics and business conventions).

More importantly, the state-imposed assessment criteria do not necessarily relate to ethical considerations but may be preoccupied with ideological concerns. The abstract and decontextualized “trustworthiness” alluded to in the policy documents is not identical to the promise-keeping ability or the professional quality of a community member that should be evaluated by the community concerned according to specific business, professional, or moral contexts. The very vague concept can be stretched to associate standards for integrity with other officially endorsed values. With Party agencies participating in standard-setting and self-regulatory organisations themselves regulated by the SCS, ideological control may be realised in the name of integrity assessment in the targeted fields. And this would hinder the self-governance in different sectors of the civil society which has developed slowly in the past three decades thanks to the marketization process.

The extra-legal nature of integrity assessment causes difficulties in seeking legal redress. The making of rules by state or the Party falls outside the scope of administrative litigation. As an alternative, disciplinary measures taken by the self-regulatory organisations may be litigated by the affected members before civil courts. However, the legality of the state-imposed standards remains a delicate matter for a civil court to adjudicate on, especially when such standards involve inputs of the Party.


[i] For evolution of the understanding of the notion “social credit” among Chinese policy-makers and advisors, see Liu Xiaoyuan et al., Woguo Shehui Xinyong Tixi Jianshe Yanjiu [Research on the Construction of the Social Credit System in China] (Beijing: Zhishi Chanquan Chubanshe, 2016) 85-91. For an account of the historical development of social credit system in national policies from 2003 to 2011, see Liu Y et al. (2014) “An Overview of Big Data Industry in China” (December) 2 China Communications 2.    

[ii] See Tufu Remote Sensing Tech Co. v. National Bureau for Surveying (1st Intermediate Court of Beijing Municipality, 27 September 2017).  

[iii] See Lin Ruzu v. Administration of Industry and Commerce of Wuhan City (Jiang’an District Court of Wuhan City, 24 November 2016).  

[iv] See Yuanzhi Toy Co. v. Market Regulator of Binjiang District, Hangzhou City (Intermediate Court of Hangzhou City, 1 March 2018).  

[v] See Lianfa Construction Co. v. Housing and Urban Planning Bureau of Sanming City (Intermediate Court of Sanming City, 16 January 2017).