The Implementation Mechanism and Governance Path of Corruption: A Transaction Cost Politics Perspective

Based on the perspective of transaction cost politics, corruption is essentially a political transaction process, and the development of complex transaction networks stems to a certain extent from the protection provided by the informal rules of corrupt transactions. These corrupt implementation mechanisms include first-party governance based on normative and trust, mutual implementation based on stable and constraint relations, independent third-party implementation for agency fees, and so on. Diversification of implementation mechanisms can meet the needs of different types of corruption transactions. For example, small and accidental corruption can be completed by first-party implementation, organized third-party implementation protects a high-frequency, large system of corruption. From the transaction cost politics perspective, the anti-corruption program must be holistic to break the path dependence of corruption, which can achieve a multi-level fundamental change from crushing the corrupt implementation mechanism and changing the social expectations to establish the authority of formal rules.


Introduction
Corruption is inter-temporal for human beings. Corruption persists in various forms in societies at different times and different levels of economic development and political systems (Erie, 2019). Ubiquitous corruption brings complex and diverse effects, and in the long run, the deterioration of this political ecology causes economic waste, political instability, and reduced governmental governance capacity. Precisely because of its pervasive existence and multiple impacts on society, corruption has become an issue of sustained multidisciplinary attention in academia. In a nutshell, scholars believe that the deficiencies in institutional structure and arrangement are the major reasons for the prevalence of corruption (Aidt, 2003). Such defects are mainly manifested in the following aspects: 1) government intervention and control of the market give officials more opportunities to seek rents and increase the space for corruption (Nye, 1967); 2) the failure of governmental power to supervise may make "decentralized corruption" gradually change into "centralized corruption"; 3) the imbalance of a strong state and a weak society causes corruption and brings about more serious bureaucracy; 4) the lack of a regular mechanism for increasing the salary level of public officials induces them to engage in corrupt activities (Graeff, 2004). The explanation of the individual motivation for corruption can be attributed to two approaches: the political economy perspective and the sociocultural perspective.
The political economy perspective, based on the rational calculating individual, usually uses the principal-agent model and views corruption as the act of government officials providing illegal benefits to their clients in violation of the rules set by the principal or the law based on their cost-benefit calculations (Schweitzer, 2004), which means that corruption is the result of the subject's choice under institutional and normative constraints (Porta et al., 2014). In contrast, the socio-cultural perspective argues that the reduction of human beings to unconstrained metazoan individuals is inappropriate and that explaining the occurrence of corruption in terms of individual rationality, especially material interests, is too simplistic and reductive. The sociocultural perspective argues that norms, rules, values, culture, and traditions in society have an important influence on individuals' behavioral choices, and that these factors internalize the intrinsic values of actors and constrain their behavior (Xiao, 2004), or that it is the internalized values that form the "moral costs" of actors, in other words, corruption is a phenomenon of moral or responsibility deficit.
There is no doubt that academic studies on corruption are important for our understanding of the occurrence of corruption in terms of formal institutional weaknesses and individual motivational mechanisms, and the proposed paths of corruption governance are mostly of practical value. However, both approaches ignore the transaction process of corruption, which makes it difficult to effectively answer an important question: how can corruption transactions be realized? Considering that transactions are subject to both legal and economic risks, how can corruption be successfully implemented and how are the various institutional elements organically integrated and intrinsically linked? What core mechanisms enable different types of corruption? And how do the implementation mechanisms influence the outcomes of corrupt behavior? Since existing studies cannot systematically answer these questions, this paper uses the framework of transaction cost politics as a basis to explore the internal logical me-

Research Framework
It is generally believed that the concept of "political transaction costs" was introduced by Douglas C. North. In North's view, there are transaction costs in political systems and organizations, including the resources consumed in the exchange of rights in the political sphere (North, 1984). Transaction cost political theory examines political transactions in the context of political contracts and argues that similar to economic transactions, political transactions are subject to contractual problems such as limited rationality, information asymmetry, opportunism, and asset specificity. However, compared to economic markets, "political markets tend to be more inefficient" and high transaction costs are responsible for the complexity of political problems (Schramm & Taube, 2003). The focus of transaction cost political theory on the problem of government failure provides a very useful perspective for discussing the process of corruption.
According to the specific framework of transaction cost politics proposed by the famous American economist Dixit (2009), corrupt behavior can be described in the following four ways. 1) Contracts as the unit of analysis. To avoid illegal behavior, the parties to a corrupt contract are rarely two known contracting parties, and the content of the contract is "hidden", even with many arrangements to avoid liability and transfer fault; 2) the enforcement of the contract. Because of the illegitimate and private nature of corrupt contracts, many political commitments are not subject to any external enforcement mechanisms, and these contracts are often enforced through informal norms and trust, repetition and relationships, and principal-agent mechanisms; 3) public agency. In corrupt transactions, where agency relationships are extremely complex and there may be several players trying to influence the behavior of public officials, there is always ambiguity about who is who's an agent. This ambiguity will severely weaken the degree of incentive of the agent, and this weakened incentive may be key to understanding some of the complex outcomes of the corruption process; and 4) limited rationality. The complexity and uncertainty of corruption contracts are more prominent than economic contracts, and participants do not have the full ability to calculate the full consequences of all courses of action, hence the importance of enforcement mechanisms (ex-post dispute resolution mechanisms) for corrupt behavior (Lambsdorff, 2002). Such enforcement mechanisms refer to the informal rules and their operation that are constructed by the parties to a corrupt transaction to reduce the cost of political transactions and facilitate the implementation of political contracts in the absence of coercive mechanisms.
Empirically, such implementation mechanisms may be multi-level and multi-dimensional and this constitutes the logical starting point of the theoretical analysis in this paper. To explain the dynamics and complexity of the differentiated corruption process, it is necessary to construct a unified theoretical framework, which has been found that excessive transaction costs can induce micro-corruption among grassroots public officials (Graeff, 2004). This framework can explain the implementation process of corrupt transactions in the political market, as well as the operation of different implementation mechanisms in differentiated types of corrupt transactions.

Corruption as a Transaction
Although corruption is a fundamental topic in social science theory, there is no agreement on the definition of the concept of corruption in different disciplines.
In general, political science believes that the essence of corruption is the abuse of power and the use of power for personal gain (Lambsdorff, 2002). Sociology believes that the growth and deterioration of corruption are directly related to the development of civil society and can be defined as "the alienation of public power based on social capital and carried by corruption networks" (Oldenburg, 1987). From the economic perspective, corruption is a kind of rent-seeking behavior, which refers to the efforts of individuals or enterprises to obtain self-interested benefits in the political sphere based on the cost-benefit calculation (Husted, 1994). Transaction cost political theory applies the logic of economics to the study of political issues, providing a more microscopic and multifaceted analytical framework. In this theory, "politicians" are also self-interested, rational utility maximizers, and corruption refers to a particular contract negotiated between a bribe giver and a public official for the exchange of power and money (Williamson, 1999). Thus, in corrupt transactions, transaction costs become the key variable affecting the power arrangement. The issue of transaction costs in the economic sphere includes a range of information, negotiation, and implementation issues. When placed in the context of political markets, this theory allows for an exploration of the properties of corrupt transactions themselves and their impact on political transaction costs, allowing the relational structure and implementation mechanisms of corrupt transactions to be better understood and linked together.

Transactional Properties of Corruption
Compared with legal transactions, corrupt transactions have unique essential attributes, namely illegality, secrecy, and interlocking (Giles, 2010). Among them, the illegality of corrupt transactions can also be called non-legally enforceable, which means that the "contractual obligations" of corrupt transactions are not protected by law and cannot be arranged or enforced through the state as a third-party mechanism. Even in societies where corruption is prevalent, the risk of prosecution for corruption still exists, so corrupt transactions are usually confined to the private sphere of the actors and are not public. Secrecy, low transparency, severely restricted participation, and high exit costs are the natural requirements of the trading environment for the illegality of corrupt transactions (Porta, 2012).
At the same time, because corruption is subject to the dual risks of external scru-tiny by the regime and internal deception in the transaction, the relationship between corrupt parties does not end with the transaction but can become interlocked to prevent extortion and profit-seeking behavior, requiring the relationship to be maintained by linking the corrupt relationship to a legitimate one or by repeating the transaction. Thus, the interlocking nature of corrupt transactions is sometimes summarized as continuity. In the context of multi-person political transactions, transaction attributes determine transaction costs, and transaction costs influence the possibilities and specific forms of transactions.

Transaction Costs of Corruption
Transaction attributes affect transaction costs, which is the central point of transaction cost analysis. As a form of transaction that is not protected by state power, corruption faces both the uncertainty of judicial investigation and punishment by the state and the threat of backstabbing and extortion by the other party, and information about the transaction may be distorted or strategically exploited to gain more rewards (Lambsdorff, 2007). The time, effort, money, and other resources invested in reducing these uncertainties, securing the transaction, and facilitating a successful transaction constitute the transaction costs of corruption. Similar to the definition of market transaction costs, given the assumptions of opportunism, limited rationality, and information asymmetry, corruption transaction costs can be divided into three dimensions according to the composition of the political transaction process: information gathering costs before a deal is struck, contract negotiation costs during the deal, and contract implementation costs after the deal (Porta, 2012).
The first is the information-gathering cost of the trading partner. The first thing that a corrupt deal faces is finding a trading partner. Both demand-based corruptions for the search for willing private bribes and supply-based corruption for the identification of willing officials require sufficient information. However, information has a cost, and the cost of information gathering in the corruption field is higher relative to the cost of legitimate transactions. Because of the illegal nature of corruption transactions, information is scarce, asymmetric, and easily distorted, and collection activities can affect and threaten the individual reputation and career development, so information collection must be both discriminatory in terms of authenticity and confidential throughout the process. In the search for corrupt trading partners, a wide range of information is required, including the resources paid and received by the bribe or bribe-taker, the identification of potential trading partners, and the reliability of the trading partners.
The relevant information is also collected in multiple ways, including direct questioning, indirect probing, and private and confidential dissemination of anti-corruption information. Some scholars argue that the factors that influence the way corrupt trading partners are searched, and the transaction costs include the level of trust, information channels, cultural or ideological differences, geographical distance, and the general level of regional corruption (Vannucci, Open Journal of Social Sciences 2009). How contacts are established, and corrupt offers are expressed is critical to the conclusion of a corrupt transaction, not only affecting the establishment of the transaction relationship but also determining the magnitude of the risk of being reported to the parties involved.
The second is the cost of negotiating a corrupt transaction. Corrupt transactions are "big bucks for little money" deals for all parties, but they do not come about naturally; rather, they are a complex bargaining process that incurs diverse negotiation costs. These include the opportunity cost of the negotiator's time, the cost of delay, or the cost of a breakdown in negotiations (Donatella & Alberto, 2005). For reasons of confidentiality, the terms of a corrupt agreement are generally not likely to be formally contracted in writing or even discussed in a public manner, which means that the negotiation costs of corruption are higher than average. In the negotiation process, the core issues mainly concern the amount of remuneration, the method of payment, and the preferential treatment available to the private party. Since the negotiating power of corrupt parties is unequal, bargaining power is related to the asset specificity of the official's power, the time priority of the actors, the discount rate, the number of alternatives, and other factors (Cartier-Bresson, 1997). At the same time, the cost of negotiation is influenced by both the content of the negotiation and the number of personnel. In general, the definition of the disputed issues in the transaction contract is the key for the parties to reach an agreement, and the efficiency of the definition will directly affect the size of the negotiation cost; in addition, the negotiation cost is proportional to the number of participants, and the increase in the number of participants is likely to make it more difficult to reach a mutually beneficial contract.
Finally, there is the cost of implementing corrupt contracts. Corrupt transactions are often non-synchronous and rely mainly on verbal commitments. It is difficult to reach self-fulfilling contracts or to guarantee the implementation of contracts through formal third parties (Dorney, 2010). Therefore, the establishment of additional monitoring and confidentiality mechanisms is an inevitable process for corrupt transactions (Graeff, 2004). After a corrupt contract is concluded, time, money, and other resources need to be invested to establish implementation mechanisms. The various costs incurred by implementing, supervising implementation, and deterring and punishing partners' deceptions constitute the so-called implementation costs of corrupt contracts. The issues involved in implementation are mainly supervising the compliance of trading partners with their commitments, identifying defecting partners, discouraging or punishing deceitful parties, and avoiding investigation and prosecution by judicial authorities. In the implementation process, the level of trust is a key factor in the level of implementation costs, and a trusting relationship between the parties can reduce the investment of time and money in the implementation phase. The cost of punishment and retaliation for betrayal is high for corrupt practices, even more, expensive than the benefits stolen by corruption.
In sum, as an illicit transaction, corruption faces the twin uncertainties of external scrutiny and internal betrayal, and corrupt transactions entail extra costs to secure the implementation of this particular transaction, with unusually high transaction costs. In this way, corrupt transactions do not appear to be necessary for implementation. However, corruption is not only not disappearing, but also ubiquitous. The reason for this is that the parties to the transaction have established the implementation mechanism and relationship structure that allows corruption to occur.

Mechanisms of Implementation of Corrupt Transactions
As mentioned earlier, corrupt transactions endogenize high transaction costs, and the mechanisms for dealing with these transaction costs affect the way parties behave. The implementation mechanism of corrupt transactions plays an important role in the political market. The complexity of political markets determines the diversity of corrupt transaction implementation mechanisms (Graf Lambsdorff, 2013). Depending on the number and relationship of the parties to the transaction, there are three types of implementation mechanisms: first-party implementation of corrupt transactions, mutual implementation, and third-party implementation (Harnay & Kirat, 2015). The choice of implementation mechanism will depend on the extent to which different inputs and outputs are observable and the differences in value between the parties to the transaction (Hellman et al., 2003). The existence of multiple intertwined implementation mechanisms enables an in-depth study of various forms of corruption and reveals how differentiated corruption implementation mechanisms address different transactions.

First-Party Implementation of Corrupt Transactions
First-party implementation refers to the internalized moral consciousness of corrupt dealers to restrain strategic behavior and comply with illegal deal agreements, and to ensure the implementation of corrupt deals through the ethical norms of "corruption with integrity" and "theft with integrity" (Lambsdorff & Teksoz, 2004). Under this mechanism, the execution of the corrupt transaction contract is a self-implementing mechanism. In other words, when corrupt behavior is internalized to a certain extent, it gradually forms informal norms in the process of social interaction, and the ethical or normative costs of violating such informal norms will make the actors feel guilty or uncomfortable. The resources that provide the intrinsic drivers for first-party implementation of corrupt transactions are norms and trust, both of which can be considered as negative social capital that drives corrupt transactions and are substitutable for each other (Leung et al., 2021). Specifically, in a given context, corruption norms have a significant impact on the behavioral expectations of both parties to a transaction. When corruption creates a social norm, this norm can dictate the behavior that actors should take in a single transaction and, in long-term transactions, enable participants to organize corrupt transactions according to needs and expectations. Similarly, trust is of great value to corrupt relationships, and trust between trading partners can lead to corrupt contracts without reference to any norms. The norms and trust that facilitate corrupt transactions may be either pre-existing or created at a later stage. In the former case, clan, ethical, political, and other social ties may reinforce first-party enforcement mechanisms; in the latter case, the negative social capital of corrupt transactions is either acquired through deliberate planning or is a byproduct of accidental actions by the parties to the transaction. Norms and trust bind actors only if participants share similar internal norms, but the effects of norms on individuals are often heterogeneous and unstable, and trust takes a long time to build. Therefore, mechanisms that rely on participants' internal self-restraint are not only situational and can only be achieved in a particular social context but are also inherently unstable and the risk of participants adopting strategic behavior is always present, which is a key limitation of first-party implementation mechanisms.

Mutual Implementation of Corrupt Transactions
Mutual enforcement of corrupt transactions is also referred to as second-party enforcement. It is essentially a mutually threatening relationship in which one party threatens to terminate a contract or retaliate with a "tit for tat" strategy to make the other party keep its promise and consciously fulfill the agreement reached, making it the natural choice of both parties to maintain a mutually beneficial relationship (Liang et al, 2015). Mutual enforcement occurs when there is a credible threat of a corrupt transaction, and corruption is transformed from anonymous to relational. The rationalization of corrupt transactions is a prerequisite for mutual enforcement to be able to work. In the view of transaction cost political science theory, corrupt transactions are socially embedded and associated with gift-giving, solidarity, and predatory power among members of society. In his discussion of the process model of corruption, Xue Gang concludes that mutual enforcement of corruption does not arise naturally but is achieved through two mechanisms: long-term or frequent transactions and the embedding of corrupt relationships in legitimate relationships (Barzel, 2002); these two mechanisms are important to ensure the reciprocity of "honest" transactions and the group enforcement mechanism of "betrayal" transactions (Liao & Liao, 2020). Repeated transactions establish stable expectations and make cooperation a more favorable strategic choice than betrayal through manipulation of rewards or retaliation (Mishra & Samuel, 2016). In addition, multiple transactions promote the spread of a culture of corruption, and a reputation for "honesty" in corrupt transactions can reduce transaction costs at all stages. When corrupt transactions are embedded in legitimate relationships, not only can the illicit nature of the transaction be better concealed, but the constraints of legitimate relationships also increase the stability and reliability of the transaction. It is found that corrupt transactions can be embedded in a wide X. F. Xu, Z. Y. Chen range of legitimate relationships; colleagues in private relationships, family, ethnicity, formal hierarchical structures, and shared values can become resources for corrupt transactions to be mutually enforced; a very complex network of relationships is formed between corrupt transaction subjects, prompting cooperation as the optimal behavioral choice (Maggio, 2021). The mutual implementation of such cooperation is based on a reliable network of social interactions and relationships between the parties to the transaction, which may be either pre-existing, such as family and ethnic groups, or intentionally established, such as friends and colleagues. However, the limitation of enforcement mechanisms that rely on mutual constraints and threats is that transactions can only occur on a small scale among acquaintances.

Third-Party Implementation of Corrupt Transactions
When the field of corruption expands, the cost of information gathering increases and the cost of monitoring agreements and partners rises, it is difficult to complete transactions through self-regulated first-party enforcement. Mutual enforcement is also difficult to function in transactions that are more anonymous and mobile. In this case, third-party implementation becomes necessary.
The so-called third-party implementation also called the intermediary participation model of corrupt transactions, is a model in which individuals or organizations that are not directly involved in corrupt transactions secure private order in corrupt transactions by selling protection services (Hellman et al., 2003). In a corrupt transaction, after mutually distrustful corrupt parties hire a third party, the third party completes the corrupt transaction by establishing contacts, coordinating negotiations, providing safeguards for the transaction, and receiving payment for these services. In effect, third parties are the brokers of corrupt transactions. The study found that third parties play a crucial role in reducing the transaction costs of corruption, and their role in corrupt transactions can be summarized as the following four points: first, to examine the trustworthiness of both parties and establish corrupt communication channels; second, to help identify corrupt partners and reduce the risk of open searches; third, to disguise bribes as commissions to give corrupt transactions a legitimate (Ni, 2001); fourth, to monitor public officials' compliance with their commitments and ensure that they are "honest bribe-takers" (Lei, 2013). Of course, third parties are by their nature agents of both the bribe giver and the bribe taker, and their reliability and incentive compatibility may become problematic, even adding new uncertainties to the transaction. But, as Barzel (2002) points out, third parties work because of their ability to help private individuals overcome transaction cost barriers and thus qualify for negotiations with the government. Both the public sector and private organizations may have this ability, the former such as political parties and local governments, and the latter such as families, corporations, and even religious organizations, who can operate corruption using potential coercion, information, economic resources, and violent means. Open Journal of Social Sciences From the above analysis, it is possible to get a glimpse of the operation mode of the corruption implementation mechanism from the perspective of different implementation subjects. This paper explains and understands the implementation mechanism of corrupt transactions by focusing on the diverse corruption phenomena, and thus establishes a typological framework of corrupt transactions based on the attributes of corrupt transactions and the perspectives of implementation agents, considering factors such as the illegality, secrecy, and interlocking nature of corrupt transactions, as well as the first-party implementation, mutual implementation, and third-party implementation patterns of corrupt transactions (Figure 1).

Differentiated Types of Corrupt Transactions
Because of the very different illegal conditions, secrecy needs, and transaction costs of corrupt transactions, which in turn evolved different adaptive types of corruption, in the political market, the essence of corruption is the use of power for personal gain, while the basic form of corruption is the exchange of political power and wealth (Song & Guo, 2016).  (Figure 2). Different types of transactions induce the input of different implementation mechanisms, and a mixture of mechanisms is even required to secure transaction implementation in specific cases.

Petty Corruption under First-Party Implementation
Petty corruption refers to the occasional, unscheduled repetition of corrupt transactions of relatively small value, such as taking and accepting favors. In general, petty corruption involves the lowest ranking "street officials" and is therefore also referred to as "street corruption". The transaction costs of petty corruption may discourage transactions due to their low value. However, it is also the small size of the subject matter, the low level of public officials, and the small number of participants that make petty corruption not only less costly to search, monitor, and bargain for, and less difficult and risky to pass on and hide funds, but also weakened moral barriers, and bribes can easily be interpreted as feelings of understanding and even gratitude to petty bureaucrats. Even with less reliable partners, petty corruption can easily generate enough trust or be perceived as an inescapable "unspoken rule" to be followed, i.e., the first-party implementation mechanism is sufficient to ensure the smooth completion of the transaction. Social ties, such as acquaintances, relatives, co-ethnics, or shared political or cultural heritage, can have a significant impact on reducing opportunistic behavior and making petty corruption transactions easier to conclude. In petty corruption transactions, the involvement of third parties is generally not required, but once third parties acting as intermediaries can capitalize on their chains of information and trust, numerous casual collaborations can be linked, and petty corruption can become sectoral or regionally prevalent. Moreover, because the risk of exposure or betrayal of petty corruption is low when simple rules of institutionalized corruption spontaneously form, they can quickly become the lubricant for smooth access to (inappropriate) public services. The prevalence of petty corruption will encourage the spread of corrupt information and rules, lower the moral constraints of society and increase the expectation of corruption, thus further lowering its transaction costs and allowing petty corruption to gradually spread into mass corruption, even forming a collective corruption at the grassroots level of "one end of a nest".

Individual Corruption under the Mixed Implementation Mechanism
Even if the frequency of transactions is not high, with the increase in the value of corrupt transactions, the anonymous transactions relying on informal norms and trust will inevitably be transformed into relational real-name transactions, and this kind of corruption is called individual corruption (Stapenhurst & Kpundeh, 1999). Public officials with higher ranks are a high-incidence group of individual corruption due to their special status and authority (Tian, 2017). Such corruption cases are not uncommon. Since the 18th National Congress, the number of officials at the provincial ministerial level and above who have fallen from power (excluding those serving in enterprises) has reached 222, causing significant losses to state property. Individual corruption is characterized by high rewards and high risks, and the trust and internalized norms between transaction subjects can effectively reduce transaction costs. In close social relationships, such as relatives, friends, and personal loyalty, it is possible to motivate individual corruption to complete transactions. However, due to the contingent nature of the transaction, the "one-shot deal" nullifies the mutual enforcement mechanism, and the initial investment in building trust between unfamiliar people is very high, so the scope for first-party enforcement is very limited. However, the high level of interest creates a "structural hole" that provides great opportunities for third parties to engage in corrupt transactions, and a broker mechanism for corrupt transactions emerges. Under this mechanism, intermediaries in corrupt transactions can transmit confidential information, establish trust relationships, coordinate the conclusion of transactions, and provide protection services for corrupt transactions, effectively assuming and mitigating the high risks of illegal transactions, and intermediaries become indispensable participants in individual corruption. In individual corruption transactions, intermediaries often operate under the cover of legitimate relationships, or rather, special professional groups have the natural advantage to act as brokers of corruption transactions, such as lawyers, brokers, real estate agents, former officials, private secretaries, relatives, and so on. They are naturally or professionally more likely to establish trusting relationships with agents of public institutions. The emergence of corrupt intermediaries, especially the professionalization and organization of intermediaries, will greatly reduce the transaction costs and uncertainty of individual corruption.

Structural Corruption under Mutual Implementation
Structural corruption means that society and institutions commit foul play to-gether and the parties are ethically aware that they are committing foul play, but they cannot be punished because there is no one directly responsible for the fault (Van Sittert, 2015). The transaction amount of this type of corruption may be small, but the transaction frequency is very high. In 2018, a serious case of structural corruption was exposed in the county of Jieyang City, Guangdong Province, which involved a total of 539 people, including 19 division-level cadres and 213 section-level cadres, and all the major leaders of 52 party departments and 13 towns in the county were involved in the case (Yang & Gong, 2020).
Structural forms of corruption have the following three distinctive features: 1) almost all activities within public power organizations revolve around the receipt of bribes; 2) public officials are basically involved in an invisible corruption network that is constructed by unwritten norms and task role assignments; and 3) corruption rules are widely disseminated, and private subjects are willing to pay bribes to obtain convenience or benefits. Corruption becomes embedded in formal organizational structures through repetitive transactions over time, creating impersonal and structured transaction processes that create an additional, almost overt "hidden order" under textual laws and formal rules and regulations. In this order, mutual enforcement becomes the most convenient, safe, low-cost, and effective enforcement mechanism for public power organizations.
In structural corruption transactions, the information and expectations of the previous behavior will be gradually aggregated into structured norms: on the one hand, the transaction contract will become more detailed, forming a set of behavioral norms including the characteristics of partners, the content of expressions, and the methods of expressions; on the other hand, repeated transactions will form trust, cooperation conditions and reputation resources of both parties, and when both parties comply with the rules of the transaction, the corrupt transaction relationship will continue or even expand. When informal norms of structural corruption are formed, not only will corruption be entrenched in public power organizations, but also social expectations of corruption will be formed, which makes it difficult for anti-corruption measures to work.

Systemic Corruption under Third-Party Enforcement
Systemic corruption is the most complex type of corrupt transaction. It is characterized by high transaction frequency and high transaction value. The corrupt behavior and profit-making forms of large transactions gradually eat up the established organizational functions through repeated transactions and systematically invade the public power system of a society, which is particularly harmful to the political system and political ecology. Compared to the previous forms of corruption, systemic corruption faces more complex problems of contract implementation, which requires addressing both the risks of free-riding, deception, betrayal, and being judicially investigated and punished due to high transaction amounts and having to face the challenges of assessing the value of assets, coordinating the participation of multiple subjects and the expertise of corruption. As such, mechanisms such as self-implementation, trust, reputation, and mutual implementation are difficult to meet the needs of systemic corruption implementation but need to be operated and maintained by third-party organizations.
When corruption is institutionalized by profit-seeking third-party organizations, more complex networks of transactions emerge and may become widely accepted by society under a combination of economic incentives and social pressures. Most seriously, systemic corruption becomes more vital and destructive when it is organized and perpetrated by specialized third-party organizations.
The reason is that, on the one hand, systemic corruption often overlaps with the organizational roles of public or private institutions and is subject to informal norms of behavior and other resources such as confidential information, illicit skills, and illicit funds, and almost all public or private agents will be involved in corrupt transactions to a greater or lesser extent; on the other hand, systemic corruption transactions carried out by third-party organizations may become market activities that, with their institutional roles protect judicial control for corruption. As such, systemic corruption is either fully embedded in the organization of public power or replaces the formal order and becomes an alternative social order and logic in practice.
Typifying and structuring corrupt transactions based on dual transactional properties is important for a deeper understanding of the diversity of corrupt transactions, the variability of corrupt relations, and the complexity of corruption governance mechanisms. The typological classification of corruption by focusing on transaction attributes and transaction processes points to opening the black box of corrupt transactions, revealing the occurrence mechanism of diverse corruption patterns, and answering how traditional petty and individual corruption develops into structural and systematic corruption and forms complex corruption relationships. This lays the foundation for us to explore more effective and targeted corruption governance paths and to construct a system of punishment and prevention of corruption.

Conclusion
In the light of the political theory of transaction costs, anti-corruption measures such as reducing government intervention and reforming administrative procedures appear to be essential, but such "efficiency-based" policies can only curb corruption in the short term. When the policy enters its "decay period," the effectiveness of the anti-corruption measures will be significantly diminished. Moral indoctrination, official accountability, and public oversight can, to some extent, change social preferences and create ethical constraints, but they also face the dilemma of being too uncertain and difficult to test in practice, making it equally difficult to cut off the path of corruption. In this context, any anti-corruption measures that rely on a single re-source have their limitations. The root causes are that corruption itself is a complex and multidimensional phenomenon, influenced by many interrelated variables. These variables affect the benefits, expectations, and social values of the transaction; and second, corruption is holistic, and corruption in one sector may be the result of malfeasance in another sector. Therefore, only by proposing holistic anti-corruption governance solutions can revolutionarily changes in the field be brought about.
Therefore, to reduce the cost of anti-corruption and change the social expectations of political behavior, the following three aspects need to be considered: first, to establish various forms of accountability systems for anti-corruption initiatives, to deal strictly with disciplinary violations, and to selectively increase the transparency of government information to create a deterrent effect on corrupt transactions. This will help break the deep-rooted practices and popular beliefs, change the social expectation that "there is no corruption anywhere" and "no corruption is impossible", and raise the awareness of bureaucratic organizations and decision-makers about the significance of improving transparency and commitment to anti-corruption. This will create a social consensus that corruption is marginal, high-risk, reprehensible, and low reward. Second, to dismantle corruption governance mechanisms through democratic participation. As an institutional design, democratic instruments are a key option for solving the anti-corruption conundrum, weakening the system of corruption and breaking the mechanisms of internal recruitment, trust diffusion, and protection from corruption. In the case of anti-corruption, the effectiveness of democracy is twofold: on the one hand, it can undermine the cooperative relationship between corrupt partners, weaken the incentive to construct a network of corrupt relationships, and prevent corruption; on the other hand, it can safe-guard rights and serve to preserve and guarantee the power of anti-corruption. Third, it ensures the authority of the monitoring system. When the monitoring system is fair and credible and the cost of its application is low, the root of the existence of corruption will be cut off. As a fundamental institutional system for the prevention of corruption, the monitoring system should protect individuals or organizations from the threat of forced corruption. To this end, the function of the supervision system should be strengthened at every step around political discipline, judicial and mass supervision to ensure the credibility and fairness of supervision. In addition, controlling the cost of the monitoring system is also the key to breaking corruption. It is necessary to simplify the procedures for individuals to use the formal rules, control the extra cost and give some legal relief, which is necessary to compress the space for corrupt transactions to survive.