The Categorization of Complex Employment Modes for Takeaway Platforms

The employment mode reflects the actual form of enterprise employment management and practitioners’ work, and the legal qualification of employment relations are often analyzed combining the employment modes of platform enterprises in the academic circles. The employment modes of takeaway platforms have been combed and classified in the current research from different angles. For example, Que [6] divided employment modes into “direct mode”, “crowdsourcing mode”, and “special distribution mode”. Zheng [7] summarized employment modes as “crowdsourcing distribution mode” and “joined distribution mode”. In fact, these types are divided according to the distribution scheme announced by takeaway platforms [8]. For another example, some studies have revealed the complicated employment forms of takeaway platforms in reality. For instance, the crowdsourcing mode can be further divided into “takeaway platforms directly recruit crowdsourced riders” and “takeaway platforms recruit crowdsourced riders through crowdsourcing service companies”. The “special distribution mode” has further evolved into more complicated “takeaway platforms recruit special distribution riders through distributors”, “network outsourcing”, and “special distribution riders are registered as privately or individually-owned business”. The deficiency of this type of research is that this classification standard reflects the actual working mode of takeaway riders and the platform employment management mode. For example, “special riders” are characterized by fixed working hours, stable income, and standardized employment management, “crowdsourced riders” are featured with the freedom of taking orders and flexible settlement, while “riders as privately or individually-owned business” reflect the special settings of platforms for special riders. As a result, the classification results better reflect the cooperative relation between platform enterprises and takeaway riders, and the status of various distribution partners in complex employment is not clear and sufficient. For another example, Wang divided the employment mode of organizational platforms into three types: “A mode, in which platforms directly hire labor providers”, “B mode, in which agents hire labor providers”, and “C mode, in which labor providers register and take orders independently”. Similar divisions include “direct marketing mode”, “labor dispatching mode”, “agent mode”, and “crowdsourcing mode” [9]. These classifications overcome the shortcomings that the relation between distribution partners and takeaway platforms is insufficient, but moreover, the classification standards are inconsistent and the nature of the legal relation of their employment patterns cannot be revealed. For example, the “agent model” only generally describes the most basic forms of cooperation between takeaway platforms and agents, and between agents and riders, but ignores its characteristics different from general labor outsourcing. For another example, the “crowdsourcing mode” is divided according to the working mode and working characteristics of takeaway riders, while other modes are divided according to the relation between takeaway platforms and distribution partners. In reality, takeaway platforms have already developed more complicated situations such as “recruiting crowdsourced riders through flexible employment service providers”, which is difficult to generalize with the “crowdsourcing model”. In addition, some foreign scholars argue that the regulation of platform employment should not be limited to the routine of “contract characterization-applicable rules”, and the legal application method should be shifted to assigning responsibilities according to different situations [10]. The author thinks that it is necessary to sort out and summarize the types of employment patterns again, so as to lay the foundation for the subsequent analysis of the legal structure.

Legal Relations in the Complex Employment of Takeaway Platforms

The legal relation between the platform-related employment subjects and takeaway riders have been extracted and generalized in the academic circles to varying degrees through the complex employment mode of platforms. The main viewpoints include “labor relations”, “labor dispatching relations”, “contracts for work relations”, “labor outsourcing relations”, “crowdsourcing relation”, and “intermediary relations” [11]. Among them, the “intermediary relation theory” is an excuse for the operators of takeaway platforms to reduce their own responsibilities, and it is the alienation of legal relations [7]. The author thinks that the relation between platform enterprises and first-level distribution partners should not be summarized by “labor outsourcing relation”, and the relation between distribution partners at all levels should not be summarized using “transfer of contract” or “subcontracting”, because generally speaking, the contracting unit of labor outsourcing does not participate in the command and management of labor providers, which is inconsistent with the fact that takeaway platform enterprises still partially participate in the command and management of riders after contracting and may easily confuse the relation between multi-party employment subjects on the platform. In addition, “crowdsourcing relation” itself is not a legal term. At best, it can only be used to summarize the business operation mode of takeaway platforms but not to define the specific legal relations.

In some studies, the legal relations of complex platform employment modes have been analyzed in a segmented way. For instance, it has been pointed out that the contract structure of organizational-type platform employment modes is the combination of “outsourcing contracts” between platforms and agents and “labor contracts” between agents and labor providers [12]. For another example, some scholars believe that the legal relations between platform enterprises and labor providers is either labor relation, incomplete labor relation, and civil relation [13], or a mixture of contract for work and labor contracts [9]. Some scholars have summarized the complex employment relation on platforms as “a triangular employment relation on platforms” [4]. The advantage of these studies is that they break through the previous position of studying “single contract” and recognize the reality that the legal relation between platform-related employers is not homogenous. However, the shortcomings are also obvious: First, the legal relation between many employment subjects in the platform employment chain has not been accurately positioned, making it impossible to find the corresponding legal rules. Second, although the research on the “triangular employment relation on platforms” summarizes the legal relation of pairwise combination as a whole, it remains at the explanation from the perspective of labor subordination, and no attention is paid to the analysis path outside the labor law. And the conclusions inevitably fall into the nest of “fake labor service and real employment” again.

Imbalanced Relation between the Protection of Labor Rights and Interests and the Development of Platform Economy

Since 2018, the Supreme People’s Court has repeatedly highlighted that labor dispute trials should implement the concept of “double protection”, that is, protecting the legitimate rights and interests of laborers and protecting the benign development of platform economy to promote mutual progress. However, the policy-oriented judgment thinking has led to the loss of objectivity in some trials, and the relation between the protection of labor rights and interests and the development of platform economy is out of balance. Around 2021, the government’s attitude towards the employment of the new business platforms changed from “relatively tolerant” to “strengthening the labor regulation of platforms”. Especially after the Notice on the Joint Release of the Third Batch of Typical Cases of Labor and Personnel Disputes (No.36 [2023] of the Ministry of Human Resources and Social Security) was issued by the Ministry of Human Resources and Social Security and the Supreme People’s Court, it has been highlighted by the Court that takeaway platform enterprises should take more responsibility for the labor protection of riders and strengthen the realizability of compensation for work-related injuries or third-party infringement of riders. At the same time, the Court begins to lay the emphasis on practicing the judgment concept of “penetrating supervision” and to determine the nature of employment according to the actual performance of the contract between the rider and the employer [14]. The above-mentioned policy-oriented judgment thinking has aroused concerns in the academic and practical circles. Some scholars have pointed out that the provision of “not completely conforming to the situation of establishing labor relations” in the Guiding Opinions may alienate the original intention of “double protection” into an embarrassing situation of “double-loss” for both laborers and platform enterprises[15]; some scholars also believe that the idea of hearing labor relation disputes with “penetrating supervision” is worthy of reflection; some judges have also pointed out that there is a problem of “generalized identification of labor relations” in the trial of new employment cases. Generalized identification of labor relations will increase the cost of platform enterprises, which is not conducive to social development in the long run [16]. From the point of view of the author, these doubts are not unreasonable, and the implementation of the “double protection” concept needs comprehensive and meticulous legal norms. The reason why the complexity of the employment modes of takeaway platforms can achieve such a remarkable “legal isolation” effect is that it is not so much the complex employment subject of platforms that is “isolated” but rather the simple and single judgment criteria for labor relations. Therefore, it is difficult to find the balance point between the rights and interests of laborers and the development of platform economy only by the provisions of labor policy, and it is also difficult to coordinate the relations of rights and obligations between laborers and employers.

Lack of Clear Legal Norms and Unified Judicial Logic

Judging from the 70 Judicial judgments searched by the author from 2020 to 2023 in Table 1. and Table 2., the legal basis for the court to judge the employment dispute cases of takeaway platforms is mainly divided into two types. The first type is the rule of labor laws, such as the Labor Law and Labor Contract Law, which have not clearly defined the concept of labor relations or given definite identification standards. In practice, labor relations have been identified by courts still following the idea specified in the Notice on Establishing Labor Relations (No.12 issued by the Ministry of Labor and Social Affairs [2005]) (hereinafter referred to as the “Notice”). The criteria for judging the “factual labor relations” stipulated in the Notice have been widely criticized because the scope of identification is too broad and it cannot accurately cope with the flexible employment mode on the Internet. Moreover, the Notice does not belong to the legal category in the normative sense, so it is really helpless for the court to directly quote and apply it as a normative document.

The second is the rule of the employer’s liability for damage stipulated in Article 1191 of the Civil Code. There is controversy in the academic community regarding the premise for determining this clause. Specifically, how is the employment relationship between “employers” and “staff” defined? Is it limited to labor relations only? As a “quasi-subordinated labor service relation”, this type of relationship is not within the scope of the current labor law and civil law adjustment and thus cannot be clearly explained. Secondly, how to identify the behavior of “executing work tasks”? This is also a difficult problem in theory and practice. The behavior of “executing work tasks” is a flexible concept, and in specific case judgments, multiple factors need to be integrated to make judgments.

Fig.1.

Schematic diagram I of the judgment thinking on complex employment of takeaway platforms: disagreement on “qualification of legal relations”.

Fig. 2.

Schematic diagram II of the judgment thinking on complex employment of takeaway platforms: disagreement on “division of responsibilities for employment subjects”.

The current divergence in the judgment thinking of courts lies in two major problems: qualification of legal relations involved in the complex employment modes of takeaway platforms and the division of tort liability between employment subjects. In terms of trial logic, most courts tend to quote the rule of labor laws instead of the rule of civil laws to qualify legal relations. In addition, the priority is given to the subordinate standard of labor relations instead of observing the business logic and contract arrangement of platforms.

In the application of civil law rules, almost all courts will directly apply the employer vicarious liability rule, and few courts will combine the type and nature of the actual employment contract to determine the fact.In the labor dispute cases of the takeaway platform, this point is particularly obvious. In such cases, most courts adopt the criterion of Subordination standard for labor relations to determine the employment relationship. There are still a few courts that even abandon the existing laws, policies, and academic basis, and directly include factors unrelated to the determination of labor relations, such as personal injury or property damage, disability level, and so on (see Fig. 1. and Fig. 2.).

In traffic accident tort compensation cases, the court usually interprets the relationship between “employer” and “employee” in the employer vicarious liability rules in a broad sense, that is, any relationship reflecting the domination and being dominated, the use and being used between the unit and the individual can be identified as the relationship between “employer” and “employee”. This relationship can be a labor relationship, labor service relationship and employment relationship. (see Fig. 1. and Fig. 2.)

At the same time, the court usually combines factors such as time, location, control power, benefits, and behavioral appearance to determine whether the takeaway rider “performs work tasks”. The most frequently used of these is the Control power element standard (see Fig. 2.). In court decisions, this standard is usually manifested as the supervision and management of riders by platforms or distribution partners, and “supervision” and “management” are the core meanings of “personality subordination” in labor subordination. It can be seen that whether it is a labor dispute case or a tort compensation case, the court gives priority to the subordination standard for labor rather than the observation platform business logic and contractual arrangement.

As for the division of the liability of employers, most courts follow the judgment path of determining the employer of the infringing rider first and then determining the vicarious liability of the employer. In terms of the judgment results, most courts held that there was no direct employment relationship between the rider and the platform, and the distribution liability should be borne by the distribution partner who had an employment relationship with the rider, while the platform did not bear the distribution liability. Some courts hold that there is no employment relationship between the rider and the distribution partner and the platform, and the rider should bear the responsibility alone. Some courts believe that the platform should bear the responsibility for employment independently. There are also a few courts that platform companies and distribution partners should be jointly liable for compensation. (See Fig. 3. and Fig. 4.)

It is worth noting that in the cases determining the responsibility of the takeaway platform, the reasoning of most court decisions focused on demonstrating that the rider met the characteristics of the task of performing the platform. In the cases denying the responsibility of the delivery platform, the reasoning of the majority of court decisions focused on demonstrating the non-employment relationship between the platform and the rider (see Fig. 3.). Few courts can analyze the nature of the cooperative relationship between the workers or between the workers and the riders through the contract between them, and then determine the responsibility of the platform.

Fig. 3.

Distribution chart of qualitative basis for complex employment legal relationships on takeaway platforms in court adjudication.

Fig. 4.

Distribution chart of takeaway platforms responsibility subject type in court adjudication.

Lack of Normalization and Stability in Referee Interpretation Methods

The omission of legal norms will inevitably lead to the lack of normalization and stability in the interpretation method of court referees. In practice, the phenomenon of “same case with different judgments” is prominent because of the excessive discretion of courts. To meet the policy requirements inclined to protect laborers’ rights and interests, for example, courts have made expansionary explanations of different scales for the labor relation determination standards stipulated in the Notice. In this process, some interpretation techniques tend to be controversial. The first controversy is about the “control and autonomy” relation between platforms and riders [17- 19] as well as the evaluation criteria for “control power”. In the employment of platforms, the control power (or “dominance”) of the employer and the autonomy (or “freedom”) of laborers often show a trade-off relation. The control power is the main index to measure whether the laborers’ subordination establishes labor relations. Generally speaking, platforms with strong control over labor providers are of strong labor subordination, which belongs to conventional labor relations and is regulated by labor laws. The control power is not as good as that of conventional labor relations, and labor subordination is weak, which belongs to “atypical labor relations” and is adjusted by other security mechanisms [9]. However, the “strength” of “control power”, as a legal evaluation index, cannot be definitely demarcated due to the openness and fuzziness of its own definition. Even the famous “control test composed of multiple factors” in history has not really solved the classification of platform laborers in the legal sense. Therefore, introducing more specific employment factors is not the optimal solution to distinguishing between conventional labor relations and atypical labor relations. The second controversy exists between “priority of fact finding” and “priority of parties’ express consent”. At present, most courts tend to apply the principle of “priority of fact finding” to identify employment relations. The principle of priority of fact finding requires that the priority should be given to the actual performance behaviors of two parties during judicial determination of labor relations, instead of taking the contract name or type as the basis. As far as the original intention is concerned, the principle of “priority of fact finding” aims to identify hidden labor relations in platform employment and protect the legitimate rights and interests of laborers. However, there are still many scientific theory disputes about this principle itself. For example, whether the “fact” in the “priority of fact finding” refers to the fact of objective performance or the fact agreed by the parties, and the whole fact of the case or the fact of labor subordination remains inconclusive. For another example, the premise of applying the principle of priority of fact finding is to correctly apply the existing law. However, the “labor subordination” specified in the Notice is not a legal standard in a strict sense, failing to provide a solid current law basis for judging the objective performance facts. So, can “the protection of labor rights and interests takes precedence over the autonomy of private law” contained in the principle of “priority of fact finding” be applicable to all scenarios of platform employment? In practice, labor relations and civil relations are often confused with each other in platform employment. Imaging that the factual elements of a case are not only the elements to judge the standard of labor contracts but also the elements to judge the standard of specific labor service contracts, then how will the court evaluate and choose the objective facts embodied in the express consent articles with the property of non-labor contracts and the labor payment involved in the case? Therefore, the author thinks that the judgment methods advocated by the current policies are worthy of reflection. If courts always ignore the commercial logic and contract arrangement of the Internet takeaway industry and are firmly entrenched in the identification path of “subordination + element type” labor relations, the labor relation determination” of platform employment will be easily generalized.

Commercial Franchising Mode of Takeaway Platforms

The commercial franchising mode is the main employment mode adopted by takeaway platforms at present, which provides more than half of the “capacity” for takeaway platforms. Meituan and Ele.me platforms are examples of such platforms. In order to cope with the employment pressure all over the country and meet the challenges of cross-regional, multi-network, and decentralized transportation resource management, these two giants will choose to find local distribution agents to contract the distribution management rights of some areas, and the agents will recruit full-time riders in the cooperative areas to deliver orders for their own takeaway platforms. These riders are also called “special riders”. In terms of specific operations, firstly, platform enterprises and distribution agents conclude cooperation agreements, such as “city cooperation agreement”, “distribution agent cooperation agreement”, and “distribution station cooperation agreement”. The content of such agreements generally includes three items: The first is the use terms of brand authorization. Regional distribution agents have the right to use the brand service trademarks and logos authorized by platforms. The second refers to business support and supervision and management terms. On the one hand, platforms provide technical flow, training management, and account settlement support for cooperative business. On the other hand, platforms strictly control distribution agents to carry out distribution business according to the requirements of platforms by setting up training for distribution agents, management rulers for riders, rider distribution behavior rules, and penalties for agent violations. The third is the term related to platform service fees. Platform enterprises collect platform service fees or “brand cooperation commissions” from distribution agents according to a certain proportion. Secondly, commercial franchising distribution agents set up several stations in their contracted areas, and at each station, the station manager and assistant are responsible for daily organization and management of distribution capacity and the man-machine conflict coordination in the deployment of special riders based on the distribution tasks set by platforms.

It is worth noting that “algorithm” is the key technical support for the development of commercial franchising of takeaway platforms [20-21]. The “business support” clause agreed in the franchise agreement of platforms mainly refers to the distribution network, traffic and information technology support with the algorithm as the core, which is embodied in formulating the payment mechanism, task distribution time, customer rating mechanism, rating standard, and reward and punishment rules based on digital technology, monitoring the riders’ labor process, and assigning tasks, assessing and appraising, implementing rewards and punishments, and paying distribution salary according to the data results[22-23]. In this way, the online supervision of the algorithm is the main part, supplemented by the offline bureaucratic monitoring of distribution stations, and takeaway platforms realize the dual labor management of escort riders through the commercial franchising mode [24]. This also constitutes the fundamental difference between the employment mode of takeaway platforms and the traditional labor outsourcing. In the above cooperation mode, the legal relation between takeaway platforms and distribution agents conforms to the definition and requirements of commercial franchising [25-26] and is a typical commercial franchising relation. According to whether distribution agents continue to outsource the distribution business, the above-mentioned commercial franchising mode can be subdivided into the following two modes.

(1) Simple business franchising mode of takeaway platforms

The basic structure of this model is that takeaway platforms are commercially franchised and special riders are directly recruited by distribution agents (see Table 3.). Since special riders are more closely related to distribution agents under this mode, their relation will be usually examined first by courts in practice, and the relation conforming to the legal features of labor relations is identified as labor relation [27] and that not conforming to the legal features of labor relations is identified as non-labor relation. Meanwhile, the relation between takeaway platforms and distribution agents will be rarely directly identified by courts. According to individual courts, the two present a “cooperative relation” or a “legal partnership” or the content of the cooperation agreement between the two is just generally described.

In many cases, takeaway platforms claimed that the agreement was a commercial franchising contract to show that the distribution agent could bear the responsibility independently, but few courts responded positively to this. What is the legal relation between takeaway platforms and riders under this mode, and do takeaway platforms need to take responsibility for riders recruited by the distribution agent? Although the current court decisions vary widely, the Regulations on Commercial Franchise does not stipulate franchisors’ external tort liability, and the academic theory about the employment risk of takeaway platforms as franchisers is still in the discussion stage [7, 27]. But at least it should be realized that the franchisers’ substitution responsibility in commercial franchising may be another explanation path to investigate the employers’ responsibility of takeaway platforms besides the thinking of labor laws.

(2) Complex commercial franchising mode of takeaway platforms

Difficulties in recruitment, high mobility, and changeable demand for employment in business peaks and valleys are common pain points in the employment field of takeaway platforms. Therefore, the complexity of this mode is reflected in the fact that after being authorized by platforms, distribution agents once again outsource the personnel management transactional work such as rider recruitment, salary distribution, social security, and provident fund payment, and hand it over to human resources service agencies for management. Its overall structure is as follows: commercial franchise of takeaway platforms, human resources outsourcing of distribution agents, and flexible employment services provided by human resources service institutions (see Table 4.).

In practice, the agreements signed between distribution agents of takeaway platforms and human resources service institutions include the platform service agreement of flexible employment, the outsourcing contract, and the labor contract. When reviewing such employment modes, courts do not concern the agreement in the form of agreements, but they more tend to investigate the legal relations between riders and distribution agents according to the labor management fact and subordinate characteristics. Similarly, the phenomenon of “same case with different judgments” is more prominent because the standards for determining labor relations are too broad. After the policy shifted to strengthen the standardization of platform labor and employment in 2021, this kind of employment mode is more likely to be regarded as “the fact that distribution agents implement labor employment in the name of outsourcing” [28].

As far as the author is concerned, the above flexible employment mode of distribution agents is a typical kind of human resource outsourcing. In China, human resource outsourcing belongs to a subdivision type of “human resource service” in the national economic industry classification and also a professional service form recognized by the state to support flexible employment. It means that employers subcontract some positions to outsourcing units (mostly human resources service institutions), which recruit personnel according to business processes and job responsibilities, and these employees provide services to employers. Outsourcing units are responsible for personnel management of the outsourced employees and employers take charge of business management. From the perspective of legal relations and according to the relevant regulations of the Ministry of Finance, human resource outsourcing service, as the object of tax payment, belongs to “broker-agent service”. It can be seen that the flexible employment service provided by human resources service institutions belongs to agency behavior, and the consequences of agency behavior are attributable to the contract-issuing unit. Meanwhile, the employee relation of human resource outsourcing is in the employer [29], so its employment risk and responsibility should also vest in the employer. Specific to the employment modes of takeaway platforms discussed in this research, the distribution agent, which serves as both the agent and the employer, should bear the employment risks and responsibilities of riders.

The problem left here is that this mode is a combined employment mode of commercial franchising and human resource outsourcing. Once the responsibility arises, how can the responsibility be distributed between the franchiser’s takeaway platform and the franchisee’s distribution agent? In this regard, the judgments of individual courts on traffic accident liability disputes in the express distribution industry are quite enlightening. The judgment first confirmed that the franchising relation between the two parties was commercial franchise according to the Franchise (Franchise) Contract signed between the franchising company and the franchised company, and the franchisee was mainly liable for compensation. At the same time, the principle of “degree of control and degree of fault” was taken as the judgment standard for the franchiser to bear the responsibility of the employees from the franchisee’s contractor for causing harm to the third party, and the franchisor was judged to bear the supplementary compensation responsibility.

It is worth noting that there are many misunderstandings about the relation among takeaway platforms, distribution agents, and human resources service institutions, which, for example, collectively refers to “contracting” relation [4]. Others think that the relation between distribution agents and human resources service institutions is “transfer of contract” or “subcontracting” [30]. There are also individual courts that identify the relation between the three as “joint venture relation”. The theory of “joint venture relation” is exactly the same as the theory of “multi-party employment labor relation” [31-32] advocated by academic circles in recent years. In this regard, the author holds an objection. This is because, first, the fundamental feature of a partnership joint venture is that all parties in the joint venture jointly invest, operate, and share profits and share risks. In this case, Zhuhai Mesoda Company and Anhui Bodu Company have not jointly invested in Beijing Sankuai Company (“Meituan Platform”), a “joint venture” recognized by the court, let alone jointly operated and shared the profits. Secondly, there are essential differences between takeaway platforms and distribution agents and between distribution agents and human resources service institutions. First, commercial franchising requires the franchisee to maintain a high degree of consistency with the franchiser in personnel management and business management, while human resource outsourcing allows the contracting unit to hand over some personnel management functions to a third party, while retaining the management right of the core business. Second, commercial franchising usually does not allow the franchisee to “be authorized twice”. For example, in the “General Rules for Urban Cooperation” of Meituan Platform, the types and scenarios of outsourcing such as unauthorized anchor (borrowing qualification) and transfer by distribution agents are business violations. However, human resource outsourcing allows the contracting unit to continue to decompose the personnel management functions and subcontract them to different human resource management institutions. It can be seen that the operation mode of “takeaway platform-distribution agent-human resources service institution” is different, so are their legal relations, which cannot be explained by the general “labor outsourcing”, “contracting”, and “multi-employment labor relations theory”.

In addition, in reality, human resources service agencies only collect outsourcing service fees from distribution agents, but do not actually undertake corresponding labor and employment obligations. At the same time, it is not excluded that distribution agents sign labor dispatch agreements with human resources service agencies to avoid labor relations and carry out the operation of “fake outsourcing and real dispatch”. The former situation cannot be separated from the court’s use of labor facts to reveal the nature of its “fake outsourcing and real employment”, while the latter situation also requires the court to jump out of the labor law path and carefully identify the agreement between the two parties in the case and the actual performance of the contract.

Takeaway Platform Employment Model

As a supplement to the above-mentioned commercial franchising mode, the employment modes of other takeaway platforms all take crowdsourcing as the terminal labor organization mode, and their employment modes have several outstanding characteristics: on the one hand, the way of rider registration and order receiving, the workplace and time, the salary composition, and settlement are more flexible than those of special riders, so they are called “crowdsourced riders”. From the perspective of order-receiving autonomy, platforms do not have the command over crowdsourced riders. On the other hand, once receiving the order, riders need to abide by the platform rules, and meanwhile, be controlled by the platform algorithm technology, and the degree of supervision and punishment is no different from that of special riders. The tension brought by the coexistence of labor autonomy and platform strength makes the standard of identifying riders’ personality as a subordinate factor split from the inside, and the explanatory power of traditional personality as a subordinate standard also drops greatly.

In this innovative mode of platform employment, platforms and riders sign an electronic crowdsourcing service agreement through the APP on the mobile terminal (see Table 5.). There are still some new discussions on whether the nature of the agreement is subordinate labor relation, equal employment relations, labor service relation, or contracts for work relations, and whether laborers belong to employees or independent contract laborers. The reason is that the tort liability of employees and labor providers differs from that of laborers, such as tort liability, social insurance benefits, and labor environment protection. Labor providers evidently vary in interest relations.

According to some local regulations, “crowdsourced riders are flexible employees who establish legal relations such as labor service and contracts for work with ‘platform enterprises’, without no labor relations, part-time employment relations, or actual employment relation”; in the early stage, some courts had once determined that the two had labor relations, where platforms assumed the tort liability of the employer; some courts also hold that the labor management intensity of takeaway platforms for crowdsourced riders cannot reach the standard of labor subordination[28], and besides certain economic subordination, the personality subordination and organizational subordination between the two parties are weaker than the standard labor relations, so it is impossible to identify labor relations; some courts found that the two belonged to labor service relations, cited the provisions in Article 11 of the Interpretation of the Supreme People’s Court on Several Issues Concerning the Application of Laws in the Trial of Personal Injury Compensation Cases in 2003, and sentenced that the takeaway platform should bear the employer’s responsibility for the rider’s personal injury; some courts avoided the identification of legal relations, directly applied the rules of tort liability of employers, and required the platform as an employer to bear the liability for damages caused by crowdsourced riders. According to some courts, the contract directly signed by platform enterprises and crowdsourced riders established a contract for work relation. When the rider causes damage to people, the contractor, that is, the rider, shall bear the responsibility. As the ordering party, the platform generally does not bear the responsibility for the damage caused by the rider, and the platform enterprise, as the ordering party, should just bear the corresponding liability for compensation for the fault of “instruction” or “election and appointment”.

At present, the academic understanding of crowdsourcing legal relations of takeaway platforms mainly focuses on “the theory of the third-type labor employment relation” [33-36], it is considered that the “platform-outsourcing” legal relation is a transitional form between labor relation and civil relation, and crowdsourced riders should be protected by society based on the “economic subordination” of their takeaway platforms or platform cooperative enterprises. According to “the theory of type fusion contract”, the employment of outsourcing-type platforms embodies the mixed feature of contracts for work and labor contracts, which should be respectively applicable to the contract law and the labor law [9].

On the other hand, there are some shortcomings in the above viewpoints. The “contracts for work relation theory”, for example, is not consistent with the actual situation of employment on takeaway platforms. Because as far as the typical contracts for work relation is concerned, the contractor is independent in providing services, that is, the contractor does not have to obey the ordering party in the details of the work. Therefore, if the contractor causes damage to a third party or damages itself due to its work, the ordering party shall not be held liable [37]. The delivery service of crowdsourced riders is always under the technical control of takeaway platforms, which is obviously inconsistent with contracts for work relations. Moreover, if the legal relation is determined according to contracts for work, crowdsourced riders will bear the losses independently, which obviously hinders the realization of the policy goal of protecting riders. Taking “the third-type labor employment relation” as an example, the theory argues that the platform crowdsourcing model is characterized by “no personality subordination but with economic subordination” [38]. The author believes that this understanding is biased. This is because, if the criteria for determining the subordination of labor relations are taken for contrast [39], the three elements that reflect the subordination of crowdsourced riders’ personality, namely the command, supervision, and punishment of takeaway platforms, have not completely disappeared, especially the command and supervision of takeaway platforms for riders by using algorithm technology is not weaker than that of the platform for special riders. Similarly, the “theory of type fusion contract” holds that the “autonomy of labor providers” in this mode marks the weakening of the platforms’ control over the rider’s labor and the decline of the rider’s personality subordination [9]. However, the author thinks that the “autonomy” of crowdsourced riders is more manifested. Crowdsourced riders are not subject to the bureaucratic management of offline distribution stations, and enjoy the freedom to take orders and the freedom to choose distribution tools and working time and place, but it does not mean the freedom of labor after the riders take orders, and their distribution process is always monitored by takeaway platforms. Crowdsourced riders will face fines or credit downgrade if they violate the platform regulations. What’s more, the autonomy of crowdsourced riders may also be constrained by a series of “scoring” and “level” rules set by platforms. Specially, what most crowdsourced riders can choose are the delivery orders left by special riders from remote locations and relatively low unit prices. In order to effectively send orders, the Meituan takeaway platform has set a number of “crowdsourcing levels” according to the effective completion rate and on-time delivery rate of orders. Only when riders reach a certain level can they enjoy the privilege of high-quality orders that are close to each other and cancel orders for free. At the same time, the Meituan platform will also reduce the level of riders who refuse to receive orders continuously. The crowdsourced riders refusing to receive orders on Ele.me platform will be restricted by the platform in order-receiving. If the restriction continues, the riders will be subject to permanent account ban by the platform. To sum up, although the intensity of economic subordination and organizational subordination of crowdsourced riders the platforms under the “platform-crowdsourcing” model cannot meet the standard of standard labor relations, the model generally presents the characteristics of “strong platform control” and “low rider dependence”, which meets the conditions for establishing civil employment relations.

Labor Dispatching Mode of Takeaway Platforms

In this mode, crowdsourced riders register on takeaway platforms and sign the Network Contract Agreement with human resources service companies online through takeaway platforms. The takeaway platforms and the human resources service companies sign Platform Service Agreement, Outsourcing Service Agreement, and Service Cooperation Contract. Superficially, the employment mode is quite similar to the “distribution agent-human resources service organization-special rider” structure under the commercial franchising mode. In fact, the legal structure of this employment mode is different from that of human resources outsourcing, which is more in line with the legal representation of labor dispatch (see Table 6.).

First of all, these human resources service agencies usually have the qualification of labor dispatch, and most courts have determined that human resources service companies and riders in this mode present labor relations after substantive examination. Secondly, for platforms, the distribution service provided by crowdsourced riders belongs to a kind of transportation capacity supplement and an auxiliary job position. Thirdly, these human resources service agencies are only responsible for recruiting crowdsourced riders, providing riders with clothing and business training, paying salaries, and paying social insurance premium, etc., and do not actually participate in employment management, while platforms directly direct and manage the distribution process of crowdsourced riders by using algorithm technology, rules, and regulations. In this labor dispatching mode, human resources service companies are the labor dispatch units and takeaway platforms are the employers, accompanied by the labor dispatch contract relation between the human resources service company and the takeaway platform, the labor contract relation between the human resources service company and the dispatched crowdsourced rider, and the actual employment relation between the takeaway platform and the crowdsourced rider. In this case, if the crowdsourced riders themselves are damaged, in principle, the human resources service companies should bear the employer’s responsibility. If the crowdsourced rider causes damage to a third party due to the delivery service, the takeaway platforms shall bear the tort liability according to the second paragraph of Article 1191 of the Civil Code, and the human resources service companies shall bear their corresponding responsibilities if they are at fault.

In practice, the legal relation determination by courts under this mode is not consistent. Some courts only examine the relation between human resources service companies and crowdsourced riders, and usually conclude that they have established a labor relation or an employment relation. In some “typical cases”, the courts judge that they do not constitute a labor relation according to the agreement signed between the takeaway rider and the platform company and their actual performance. As a result, crowdsourced riders have to bear the loss of wages induced by their own damage or the deduction by human resources service companies during the distribution process. The legal relation between the takeaway platform and the rider in this mode has not been clearly identified by most courts. The author believes that there are two reasons why courts adopt the above-mentioned adjudication strategy. First, there is a lack of overall analysis of the employment mode of “platform-human resources company-crowdsourced rider”, and it is impossible to recognize the essence of the legal structure of this employment mode. Second, a deviation exists in the understanding of the classification standard for the complex employment modes of takeaway platforms. The employment relation involving special riders is more likely to be identified as a labor relation, while the employment relation involving crowdsourced riders tend more to be identified as a non-standard labor relation.