Good Governance within Public Participation and National Audit for Reducing Corruption

Abstract

The globalization process and acceptance of a new paradigm for world development aimed at attaining sustainable development require extending the concept of good governance, which allows the improvement of institutional quality. At the same time, governance performance depends on corruption and public participation in national audits. The paper aimed to analyze the impact of public participation in national audits on corruption governance. The object of investigation was 30 Chinese provinces for 2008–2017. The study applied a regression model with the preventive, exposed, and defensive functions of audits. The findings showed that the “governance view” of the audit is the root cause of the audit. In addition, the “immune system” function of national audits has an inhibitory effect on corruption, among which the deterrent effect of the defense function is the most-influential. The influence of public participation in national audits on corruption governance is mainly reflected in the deterrence and resistance function of national audits. The paper contributes to audit theory by incorporating the public participation mechanism, improving the governance path of “national audit—public participation—corruption governance”. It allows improving the national audit “immune system” function by considering the democratic connotation of decision-making and developing a good governance environment.

1. Introduction

In the past 1000 years, China’s GDP per capita has remained almost unchanged, and vertical rises have occurred since 1978. Thus, snowballing economic growth [1] (GDP per capita increased from CNY 364.5 billion in 1978 to CNY 827,122 billion in 2017) led China to become the world’s second-largest economy. In addition, rapid economic development has intensified corruption issues in China. In October 2014, the Fourth Plenary Session of the Eighteenth Communist Party of China declared “fully implementing the rule of law”, adhering to strict law enforcement, strengthening the restriction and supervision of power, incorporating power into the system of the cage of regulation, widening the channel of mass supervision, and improving the complaint system, in accordance with the law to safeguard the people to supervise the power of state organs [2]. The report to the 19th National Congress of the Communist Party of China stated that “The Thought on Socialism with Chinese Characteristics for a New Era must adhere to the thought of people-centered development”, which is also the basic value orientation of comprehensively deepening reform [3]. The major theoretical achievement of the people-centered development philosophy is to incorporate them into the national governance system [4], improve the administrative decision-making mechanism combining “public participation [5], expert demonstration and government decisions”, enhance the credibility and execution of decision-making, and promote the modernization of the governance system and capacity [6]. Jiayi L. [7] outlined that the national audit has the function of the “immune system” of preventing, revealing, and resisting, and it is an important part of national governance. However, the state audit supervision force is limited, and citizen participation can be an unlimited social public opinion supervision force [8,9].

National auditing systems are an important component of good governance and play a critical role in promoting the sustainability of current social and economic systems [10]. These systems help to ensure transparency and accountability in the management of public resources, which is essential for promoting trust and confidence in government institutions. By providing independent verification and oversight of financial and operational processes, national auditing systems can help to prevent and detect fraud, corruption, and other forms of misconduct that can undermine the stability and integrity of social and economic systems [11,12,13,14]. Considering audit theory [15], motivation is an important dimension of audit performance. The results of the analysis of the audit theoretical framework show that the audit theory circle includes the concept of fiduciary responsibility [10], the concept of the separation of two rights [11], the principal agent concept [13,14], conflicts of interest [16], information [17,18,19], insurance [20,21], the shadow economy and institutional quality [22,23,24,25,26], behavior [27,28,29], supervision [30], and the concept of maintaining property rights and national governance [31,32,33]. Scholars [34,35,36,37] outline that with “people’s direct participation and self-management” in the 18th Century, the national audit has been constantly improved in practice, actively exploring how to organically combine public participation with the national audit. In China, the 2004 Implementation Outline for Comprehensive Recommendation of Administration by Law proposed for the first time to “extensively listen to the opinions of all sectors of society” in major decisions. The long-term practice of public participation in national audits provides a certain basis for the effective combination of research.

However, the results showed that scholars [7,10,11,12,13,14,15,33] focus on the investigation of the internal control system, internal audit system, and contract system. At the same time, the accuracy of audits depends on considering external dimensions that affect national audit performance. This study fills the research gap in audit theory and contributes to the theoretical background by developing an approach for assessing the impact of public participation in national audits on corruption governance, in addition to the audit analyzed within the “immune system” concept.

The paper has the following structure: the Literature Review Section explores the theoretical background on audits considering the impact of public participation in national audits on corruption governance; the Materials and Methods Section describes the data, instruments, and methods to check the public participation in the national audit’s impact on corruption governance; the Results Section gives an explanation of the empirical results of the investigation; the Discussion and Conclusion Section provides a comparative analysis of the obtained results with the previous studies, the policy implications for improving audits to reduce corruption, the limitations, and further directions for investigation.

2. Literature Review

2.1. The Governance Role of National Audits for Corruption under the “Governance Perspective”

The results of the analysis expounded that the three pillars of modern audit systems, namely national audits, nongovernmental audits, and internal audits, are rooted in the governance perspective, among which national audits are crucial to maintaining healthy economic development [11,16]. Among all the relevant path analyses, the “immune system” theory is the best representative to point out that national audits have ingrown functions, including prevention, exposure, and resistance [7]. Corruption is a complex and multifaceted phenomenon, which involves the abuse of power, position, or authority for personal gain or other illegitimate purposes [35,38,39,40]. Rose-Ackerman S. [38] defines corruption as “the abuse of public power for private gain”. The scholar confirmed that corruption restricts the economic growth of the country and justifies the relevant anti-corruption policies. Alnasaa et al. [39] define corruption as “the misuse of public office for private gain”. They argued that corruption can hinder development by reducing investment and innovation. Rothstein B. [40] defines corruption as “the use of public power for private benefit”. He argued that corruption involves public officials or institutions using their authority or resources to advance their own interests or those of their close associates, rather than serving the broader public interest. Rothstein B. [40] emphasized the importance of institutions and social norms in preventing corruption, as well as the role of trust and social capital in fostering a culture of integrity. He studied the relationship between trust, social capital, and corruption, finding that societies with high levels of trust and social capital are less corrupt and more effective at combating corruption. Rothstein also advocated for institutional reforms that promote transparency, accountability, and impartiality, such as independent oversight agencies, open government initiatives, and whistleblower protection [40]. Klitgaard R. [35] formalized the phenomenon of corruption as follows: corruption = monopoly + discretion − accountability. This suggests that corruption tends to be more prevalent when there is a combination of monopoly power, high discretion, and weak accountability. By understanding the factors that contribute to corruption, policymakers and practitioners can develop strategies to reduce corruption and promote good governance. In light of this, national audits are expected to combat corruption by strengthening the supervision of power, downgrading information asymmetry, reducing opportunistic behaviors, and improving accountability systems [36,37,38,39,40,41,42]. Past studies [22,25] showed that corruption affects the institutional quality of the country. In an imperfect market with information asymmetry, the government’s provision of goods and services can be a monopolistic competition arena. The government holds a monopoly over certain goods and services and operates in an environment where competition is limited. As a result, rent-seekers can exploit this situation by trying to manipulate the government’s policies to gain an advantage in the market. Rent-seeking behavior is defined as the pursuit of profit through political influence rather than by creating economic value. In a monopolistic competition arena, rent-seeking can take on various forms, such as bribing officials, lobbying for favorable regulations, or manipulating the bidding process for government contracts. These rent-seekers seek to exploit the government’s power and resources for their own benefit, which can lead to corruption. To prevent corruption in the monopolistic competition arena, it is crucial to establish strong oversight mechanisms and promote transparency. This includes implementing anti-corruption policies, creating an independent judiciary, and strengthening the role of the media and civil society in monitoring government activities. Additionally, promoting competition and reducing barriers to entry can help to limit the power of rent-seekers and promote a more competitive environment. Thus, regular dissimilation governance of economic power via national audits is conducive to the process of anticorruption and the efficiency of real resource allocation [43]. The coordinated governance of national audits has created a new audit pattern in the new era [44,45,46]. All kinds of supervision systems, as described in the Decision of the CPC Central Committee on Major Issues Pertaining to Comprehensively Promoting the Rule of Law [47], share the common ground that the national audit is required to cooperate with the Discipline Inspection Commission to tackle corruption in a way different from the inspection authority and judicial authority. Meanwhile, the cooperation of audit authority, inspection authority, and judicial authority is key to increasing the efficiency of anticorruption [48,49,50]. Furthermore, national audits help to improve the effects and quality of anticorruption governance by integrating resources, exchanging and cooperating, sharing information, and utilizing the results [51]. The Report on the Work of the Government in 2014 [52] first suggested that governments at all levels should intensify audit announcement efforts. Since then, the result of national audits has exerted great influence on anti-corruption, in the way that a higher level of corruption will trigger the stronger motivation of the government to punish it with audit announcements. Brazilian scholars [53] found that the announcement of government audits exerts great effects on the governance of corruption after data analysis on the publication of government audits. The study [54] testified by the data analysis of 81 countries that the announcement of national audits has a positive impact on government performance and government transparency. Thus, the national audit announcement system should be promoted to put audits into practice and rectification, improve the oversight system of national audits, and establish a supervision system that coordinates with the national governance system and modernized governance capacity [54].

The most-common theory of national audits is the ingrown “immune system” function of national audits, namely the prevention, exposure, and resistance functions [55,56]. The study [57] analyzed the “immune system” via these three functions. The prevention function of national audits aims to deter misconduct and corruption by increasing the risk of being caught and punished. This function seeks to prevent improper behavior from happening in the first place [7,54,55,56]. Scholars [57,58,59] have outlined that the exposure function seeks to identify and expose instances of wrongdoing, fraud, or other irregularities that have already occurred. This involves identifying areas where risks of misconduct may exist and then using audits to uncover actual cases of misconduct. Considering the studies [58,60], the resistance function aims to strengthen institutional structures and processes that can resist future attempts at misconduct or corruption. This involves identifying weaknesses in systems and processes and recommending changes to strengthen them. These three functions are interconnected and interdependent. Thus, effective exposure to past misconduct can deter future misconduct and strengthen resistance to future attempts at wrongdoing [61,62]. Similarly, effective resistance to misconduct can reduce the need for exposure and prevention measures [7,54,55,56].

Considering the prior studies [59,60,61], the prevention function can deter individuals from engaging in corrupt activities in the first place. The knowledge that they could be caught and punished can serve as a strong deterrent, making it less likely that they will engage in misconduct. Effective prevention measures, therefore, can reduce the incidence of corruption. Scholars [7,54,55,56,57,58,59,60,61] have shown that the exposure function, by identifying and exposing cases of corruption, can create a culture of accountability and transparency. It can increase public awareness of the risks and costs associated with corruption and generate pressure on public officials to act ethically [61]. Exposing cases of corruption can also deter others from engaging in similar activities, serving as a warning about the consequences of being caught. In addition, the resistance function can help reduce the vulnerabilities that can lead to corruption [60]. By identifying weaknesses in institutional structures and processes and recommending changes to address them, resistance measures can help prevent corrupt activities from taking hold [62].

Based on the analysis, this study proposes the following hypothesis:

Hypothesis 1.

The “immune system” function of national audits helps to prevent corruption and achieve a well-governed nation.

Hypothesis 1a.

The national audit prevention function decreases corruption.

Hypothesis 1b.

The national-audit-exposed function decreases corruption.

Hypothesis 1c.

The national audit defensive function decreases corruption.

2.2. The Governance Role of Public Participation in National Audits for Corruption under the “Governance Perspective”

Based on the practice of long-term democratic politics, public participation in decision-making has profound theoretical and practical foundations. Scholars [63,64,65] have suggested that the public be involved directly in defensive elections and social management. In 2004, China issued the Outline for Promoting Law-based Administration in an All-round Way [66] and first proposed that major decisions should take the opinions of all sides into consideration. Under the influence of recent Chinese policies, local government and official departments at all levels successively introduced relevant laws, regulations, and policy documents to encourage public engagement. This reflects that public participation in decision-making helps to improve the credibility and influence of decisions, which has already become a regular basis of national governance in the new era.

Public participation serves as the social foundation for national audits to carry out anti-corruption tasks [67]. The growing level and quality of corruption will lead to an increase in the credibility and efficiency of corruption governance for the government. Facing public pressure, national audit authorities will be more meticulous and considerate in the process of inspection, supervision, and evaluation, especially in the field of livelihood projects, environmental projects, and media such as the Internet or WeChat that report corruption clues [68]. Against the background of Internet Plus, citizens air their voices by virtue of information-based means in the online public field, which offers technical support for the public to take part in democratic politics. A combination of the Internet and traditional methods of corruption governance will favorably contribute to precise and effective governance. Recent years have witnessed the fact that audit offices and local audit authorities have solved cases involving violations of laws and regulations owing to some clues exposed by online media such as the Internet, WeChat, and microblogs. Reporting by the public could bring national audits into full play to inspect, deter, investigate, and punish corruption behaviors. Public participation in national audits to govern corruption could improve the quality of audits. Ping Z. [69] analyzed the national audit quality chain and found that public participation in audits improved the requirements for audit quality. Shaohua et al. [68] carried out empirical research on transnational data and found that public participation in audit supervision further increases the requirements for audit quality. The scholars underlined that citizens have a significantly positive influence on the acceptance of audit projects, the readability of audit results, and the transparency of government budgets. Public participation in national audits to govern corruption contributes to rectification and implementation. Considering this, the public focuses not only on the exposed stage, but also on the implementation of the audit rectification stage. Especially in the audit of people’s livelihood, the most vocal is the people [70]. This makes people have intuitive feelings about the livelihood of the project expression, which prevents the occurrence of corruption from a side perspective. Sichang J. [71] proposed that the state audit bulletin changes from the current governmental interest orientation to both the governmental interest orientation and public interest orientation, synergistically bringing into play the dual governance effects of the state audit bulletin of power restraining power and right restraining power. The practice in recent years has proven that the audit results announcement system is an important form of public participation in the national audit of corruption governance, not only to play a deterrent effect in the national audit, but to urge the implementation of rectification and, to a certain extent, to reduce the occurrence of corruption. At the same time, if public participation in the governance strategy of national audits for corruption can be effectively used, this will help to organically combine audit supervision with public opinion and social supervision. In this way, more people will be involved to improve audit announcements, and a positive interaction will come into being.

In the 1990s, China began exploring an administrative decision-making mechanism that combines public participation, expert verification, and government decision-making to bridge the gap between economic development and social progress. This mechanism enhances the credibility and execution of administrative decisions [69,70]. The influencing factors of public participation in decision-making include the interested parties, market development and the information level [71], and government credibility, government response to the public, and government information transparency [72]. In the context of governance corruption, the influencing factors of public participation include the arrangement of the political and economic system, social and cultural environment, individual population characteristics, age, literacy level, channels, public focus, and information disclosure [73,74,75]. To effectively operate any organization, effective communication based on mutual trust is necessary [76]. Task-oriented organizations require trust relationships inside and outside to acquire resources, which differs from normal organizations [77]. This highlights the importance of government credibility in policymaking and implementation. Wang et al. [78] applied trust agglomeration to measure public participation. Scholars [79] have proposed using economic, regional, educational, and informational agglomerations and regional union members to assess the composite index of trust agglomeration. Prior studies [69,70,78,80,81] outlined that the impact of agglomerations on corruption is complex and multifaceted. While some forms of agglomeration can reduce the incentives for corruption, others can create new opportunities or pressures for corrupt activities. Therefore, it is important to carefully consider the specific context and characteristics of each agglomeration when assessing its impact on corruption. Based on this, this study proposes the following hypothesis:

Hypothesis 2.

Public participation in national audits helps to curb corruption and promote good governance.

3. Materials and Methods

3.1. Research Sample Selection and Data Sources

Explained variables: The degree of corruption (Corr) was used as an explained variable. In the papers [20,42,82], the authors used standardization indexes to evaluate the degree of corruption: the Business International Index (BI) of the World Bank and the CPI Index of Transparency International are more influential [83,84]. This study evaluated corruption by the number of duty-related crime cases with reference to related research. Higher indexes indicate more serious corruption in the districts and worse corruption governance, while lower indexes reflect the opposite.

Explanatory variables: Jiayi L. [7], Rongbing H. [55], and Shu C. [56] proposed using the concept of the “immune system” to evaluate the efficacy of national audits (Adu). To measure the preventive function of national audits (Pre), which is based on the theory of the “immune system”, three sets of data were utilized: the number of individuals transferred to judicial and discipline inspection authorities, the number of cases, and the amount of money involved in each case [56]. As the preventive function is an endogenous characteristic of deterrence, this study opted to utilize the number of individuals transferred to judicial and discipline inspection authorities and other relevant authorities as the measurement tool.

Based on Wang et al. [85], the measurement of trust agglomeration (Tcp) uses the following five indexes: economic agglomeration (Por) is measured by the ratio between the production value of industrial enterprises above a designated size and local GDP; regional agglomeration (Cpr) is measured by the ratio between the city’s population and total population; educational agglomeration (Spr) is measured by the number of students in higher education per 10,000 population in the region; the regional Internet penetration rate measures informational agglomeration (Ppr); mutual trust agglomeration (Mtr) is measured by the number of regional union members.

Control variables: Based on prior research [83,85,86] on government audits and corruption governance, this study incorporated the following control variables into the regression model: human resource (Hr) measured by local average education years, indicating the influence of human resources in the district; government size (Gov_size) measured by the ratio of financial outlay to GDP, indicating the influence of government scale; wage level (Wage) measured by the ratio of the average wage of local civil servants to that of national civil servants, indicating the influence of the wage of civil servants; openness level (Open) measured by the ratio of total trade volume to GDP, indicating the influence of local openness; government transparency (Gid) measured by the government transparency index, indicating the influence of government information publication.

Based on the accessibility and availability of data, this study used 30 provinces (including municipalities and autonomous regions) in China from 2008–2017 as the research sample. The time for analysis was limited by the available and open data for all selected variables. The data relating to audits were obtained from the China Taxation Yearbook [87], including the number of people involved in cases referred to judicial organs and discipline inspection and other organs, the number of irregularities detected by auditing authorities, and the number of adopted audit recommendations; the data related to job-related crimes were obtained from the China Procuratorial Yearbook [88] and the work reports of provincial people’s procuratorates; the data on the explanatory variables of the degree of public participation and the control variables of human resources and government size were obtained from the China Statistical Yearbook, China Labor Statistical Yearbook, China Statistical Report on Internet Development, and China Government Transparency Index Report [87,88,89,90,91]. After collation, a total of 300 samples were obtained in this study. The sample regions studied in this study did not include Tibet, Taiwan, Hong Kong, and Macau. The study applied Stata 17.0 for empirical justification of the research hypotheses. The details and descriptive statistics of the chosen variables are provided in

Table 1. Variable description.

Variables	Meaning	Symbol	N	Mean	Min	Max	SD
corruption degree	the number of duty-related crime cases	Corr	300	1210.773	128	3245	697.820
national audit preventive function	the number of people transferred to judicial authorities, discipline inspection authorities, and other authorities	Pre	300	168.567	0	5683	415.495
national-audit-exposed function	the amount of irregularities detected by auditing authorities	Exp	300	1,815,127	10,379	14,700,000	2,109,286
national audit defensive function	the number of adopted audit recommendations	Def	300	6506.553	247	27,832	5037.035
public participation in economic agglomeration	proportion of the total output value of industrial enterprises above designated size in the total output value of the region	Por	300	0.391	0.118	0.536	0.084
public participation in geographical agglomeration	proportion of urban population in total population	Cpr	300	0.547	0.291	0.896	0.132
public participation in educational agglomeration	the number of students enrolled in higher education per 10,000 population in the region	Spr	300	2465	969.100	6750	909.613
public participation in information agglomeration	regional Internet penetration rate	Ppr	300	0.420	0.060	0.780	0.160
public participation in social organizations agglomeration	the number of regional union members	Otr	300	8,968,670	767,692	30,182,542	6,365,342
human resources	local average education years	Hr	300	15.1208	8.655	26.241	3.030
government size	the ratio of financial outlay to GDP	Gov_size	300	0.233	0.087	0.627	0.099
salary level	the ratio of the average wage of local civil servants to that of national civil servants	Wage	300	1.614	0.937	3.762	0.523
openness level	the ratio of total trade volume to GDP	Open	300	0.292	0.017	1.692	0.345
government transparency	the government transparency index	Gid	300	0.580	0.237	0.859	0.135

Note: N—sample size; SD—standard deviation; Min—minimum value; Max—maximum value.

.

The standard deviation of Corr was relatively large, indicating an imbalance of corruption among different regions in China. Similarly, the standard deviation of the national audit function was large, pointing to significant differences in the function of the national audit “immune system” across provinces in China. Regarding public participation, the level of urbanization, education, and social organization varied greatly among provinces in China, leading to unbalanced development. Conversely, the proportion of industrial added value in the GDP, urbanization level, and Internet penetration rate of different regions showed less variation, suggesting relatively balanced development. For the control variables, Hr had a mean value of 15.1208 and a standard deviation of 3.030, indicating that the educational level among different regions is not significantly different. Gov_size had a mean value of 0.233 and a standard deviation of 0.099, suggesting that the government’s fiscal expenditure accounts for approximately 20% of GDP, with a decreasing gap between regions. The small standard deviation of Wage indicated that, while there is a gap between the civil servants’ wages in regions with a higher level of economic development and those with a lower level, the gap between most provinces is not very significant. The standard deviation of Open was also small, indicating that China’s export-oriented economy is more apparent, with limited differences in trade openness among different regions. The mean and standard deviation of Gid were very small, suggesting that the transparency of government information needs improvement, with little difference among provinces.

The correlation between influencing factors and the degree of corruption (

Table 2. Correlation of variables.

Variables	lnCorr	lnPre	lnExp	lnDef	Hr	Gov_Size	Wage	Open	Gid	VIF
lnCorr	1.0000									–
lnPre	−0.6217	1.0000								8.16
lnExp	−0.6774	0.5531	1.0000							7.41
lnDef	−0.7956	0.6382	0.5884	1.0000						9.92
Hr	0.0157	−0.1908	0.0717	−0.1221	1.0000					3.18
Gov_size	0.5987	0.2871	0.3330	0.3226	0.0908	1.0000				1.44
Wage	−0.2743	0.2928	0.4632	0.2307	−0.0326	−0.2443	1.0000			1.56
Open	−0.1118	0.2543	0.3833	0.1476	−0.2213	−0.3568	0.7791	1.0000		1.09
Gid	0.0120	−0.0893	0.0325	0.0076	−0.0953	−0.1355	0.1491	0.2714	1.0000	1.72

Note: VIF—variance inflation factor.

) revealed that the variables related to the national audit function were negatively correlated with the degree of corruption. Therefore, the national audit function led to a decrease in the number of cases of job-related crimes. Based on the VIF assessment, its value revealed relatively low multicollinearity among the predictor variables in this regression model, as all the VIF values were less than 2.5. However, the VIF values for explaining variables were high (>5.0). For the following analyses, all variables (national audit preventive function, national-audit-exposed function, national audit defensive function) were put into the model separately. In cases of a strong correlation between Open and Wages, removing collinear variables did not reduce the VIF values significantly. Furthermore, omitting key variables from a regression model can lead to biased results. Imagine that we want to explain weekly sales by only referring to promotions.

3.2. Research Model Design

Based on the theoretical background, the empirical models to detect the relationship between the level of corruption degree (the dependent variable) and the national audit function, including the national audit preventive function, the national-audit-exposed function, and the national audit defensive function (as the independent variables), are as follows:

Model 1:

$$lnCor{r}_{i,t}={\alpha }_0+{\alpha }_1{ln\mathrm{A}\mathrm{d}\mathrm{u}}_{i,t-1}+{\alpha }_2{Hr}_{i,t}+{\alpha }_{3}Gov\_siz{e}_{i,t}+{\alpha }_{4}Wag{e}_{i,t}+{\alpha }_{5}Ope{n}_{i,t}+{\alpha }_{6}Gi{d}_{i,t}+{\lambda }_t+{\epsilon }_{i,t}$$

(1)

Model 2:

$${lnCorr}_{i,t}={\beta }_0+{\beta }_{1}Tc{p}_{i,t}+{\beta }_2{X}_{i,t-1}\times Tc{p}_{i,t}+{\beta }_{3}H{r}_{i,t}+{\beta }_{4}Gov\_siz{e}_{i,t}+{\beta }_{5}Wag{e}_{i,t}+{\beta }_{6}Ope{n}_{i,t}+{\beta }_{7}Gi{d}_{i,t}+{\lambda }_t+{\epsilon }_{i,t}$$

(2)

where $ln\mathrm{A}\mathrm{d}\mathrm{u}$—the efficacy of national audits measured by Pre, $Exp$, and $Def$; $X$ represents the three national audit function variables (${lnPre}_{t-1}\times Tcp$; ${lnExp}_{t-1}\times Tcp$; ${lnDef}_{t-1}\times Tcp$); $i$ represents provinces; $t$ represents time; ${\alpha }_i$ is a region dummy variable, representing region fixed effects that do not vary over time; ${\lambda }_t$ is a time dummy variable, representing time fixed effects that do not vary over time and used to capture the effect of a specific time on the level of corruption; $\epsilon$ is a random disturbance term.

Model 1 investigates the national audit function as the independent variable and adds relevant control variables to establish its impact. Model 2 introduces the degree of public participation (Tcp) and the cross-term between Tcp and the national audit function to investigate whether the degree of public participation and the cross-term have an effect on the degree of corruption. This study used principal component analysis to combine public participation in economic agglomeration (Por), public participation in geographical agglomeration (Cpr), public participation in educational agglomeration (Spr), public participation in information agglomeration (Ppr), and public participation in social organization agglomeration into a composite indicator to measure the degree of public participation, which is expressed as Tcp.

According to Hypotheses 1a, 1b, and 1c, the coefficients of the national audit preventive function (lnPre), the national-audit-exposed function (lnExp), and the national audit defensive function (lnDef) of the “immune system” of the national audit were expected to be negative, indicating that the national audit can exert the function of the “immune system” to suppress corruption. According to Hypothesis 2, the higher the comprehensive index of public participation (Tcp) is, the higher the degree of public participation. In this study, the cross-coefficients of Tcp and lnPre, lnExp, and lnDef, respectively, were expected to be negative, indicating that public participation can improve the governance effect of national audits for corruption.

4. Results

The second column in

Table 3. Empirical findings for Model 1.

Variables	Preventive Function	Exposed Function	Defensive Function
lnPre	−0.0652 ***	–	–
lnExp	–	−0.0464 *	–
lnDef	–	–	−0.3568 ***
Hr	−0.0054	−0.0098	−0.0168
Gov_size	3.1062 ***	3.0213 ***	3.0232
Wage	−0.0287	−0.0016	−0.1200
Open	−0.2985	−0.1331	−0.1462
Gid	−0.1580	−0.1531	−0.1464
constant	7.5619 ***	7.1243 ***	5.2482 ***
Year	Yes	Yes	Yes
N	300	300	300
${\mathrm{R}}^2$	0.1611	0.1286	0.1374

Note: *** and * indicate that the regression coefficient is significant at the levels of 1% and 10% respectively.

shows the regression results of Model 1 for the national audit preventive function and the degree of corruption. The coefficient of lnPre was negative and significant at the 1% confidence level. This indicated that the national audit preventive function reduced the number of job-related crime cases and had a governance effect on corruption. The third column shows the regression results of the national audit exposure function and the degree of corruption. The coefficient of lnExp was negative and significant at a 10% confidence level, which indicated that the national audit exposure function reduced the number of job-related crime cases and suppressed corruption. The fourth column shows the regression results of the national audit defensive function and the degree of corruption. The coefficient of lnDef was negative and significant at the 1% confidence level, which indicated that the greater the number of adopted audit recommendations, the lower the degree of corruption and the lower the number of job-related crime cases were. Therefore, the national audit defensive function can play a role in governing corruption.

The second and third columns of Table 3 reveal the coefficients for the control variables, which were negative and significant at the 1% confidence level. This indicated that the greater the share of fiscal expenditure in GDP was, the lower the likelihood of corruption. These regression analysis outputs are related to the correlation analysis results in Table 2 (the negative correlation between the national audit function and the degree of corruption), which validates Hypothesis 1: the national audit performs its “immune system” function to curb corruption and promote good governance in the country. This suggests that national audit institutions are an important tool for reducing corruption in each region of China. In addition to statistical significance, the impact of the audit function on the degree of corruption was also considerable in the quantitative analysis. The national audit defensive function had the largest impact on the degree of corruption, with a regression coefficient of −0.3568. This implied that, for every percentage point increase in the national audit defensive function, the degree of corruption would decrease by approximately 0.3568 percentage points. The impact of the national audit preventive function on the degree of corruption was much smaller than the impact of the national audit defensive function, with a regression coefficient of −0.0652. Thus, for each percentage point increase in the national audit preventive function, the degree of corruption decreased by approximately 0.0652 percentage points. The effect of the national audit exposure function on the degree of corruption was the smallest, and its coefficient was only −0.0464. This means that, for each percentage point increase in the national audit exposure function, the degree of corruption decreased by 0.0464 percentage points.

The results of Model 2’s regression analysis are presented in

Table 4. Empirical findings for Model 2.

Variables	Preventive Function	Exposed Function	Defensive Function
${\mathrm{l}\mathrm{n}\mathrm{P}\mathrm{r}\mathrm{e}}_{\mathrm{t}-1}\times \mathrm{T}\mathrm{c}\mathrm{p}$	−0.0641 *
${\mathrm{l}\mathrm{n}\mathrm{E}\mathrm{x}\mathrm{p}}_{\mathrm{t}-1}\times \mathrm{T}\mathrm{c}\mathrm{p}$	–	−0.0154
${\mathrm{l}\mathrm{n}\mathrm{D}\mathrm{e}\mathrm{f}}_{\mathrm{t}-1}\times \mathrm{T}\mathrm{c}\mathrm{p}$	–		−0.1477 **
Tcp	−0.2736 *	−0.2459	−1.2426 **
Hr	−0.0026	−0.0147	−0.0178
Gov_size	2.9240 ***	2.8124 ***	2.8592 ***
Wage	−0.0359	−0.0668	−0.0614
Open	−0.2867	−0.2267	−0.2621
Gid	−0.1154	−0.1512	−0.1314
constant	7.5630 ***	7.6719 ***	7.7546 ***
Year	Yes	Yes	Yes
N	300	300	300
${\mathrm{R}}^2$	0.1801	0.1473	0.1866

Note: ***, **, and * indicate that the regression coefficient is significant at the levels of 1%, 5%, and 10%, respectively.

. The coefficient of ${lnPre}_{t-1}\times Tcp$ was negative (−0.0641) and significant at the 10% confidence level (the second column in Table 4), which suggested that public participation can enhance the deterrence of corruption through the national audit preventive function and, thus, improve the corruption governance effect of national audit institutions. The coefficient of ${lnExp}_{t-1}\times Tcp$ was negative (0.0154), but not significant (the third column), indicating that the degree of public participation had no impact on the corruption governance of the national-audit-exposed function and the national audit institution’s ability to reveal the extent of corruption in each region, regardless of the degree of public participation. Finally, the coefficient of ${lnDef}_{t-1}\times Tcp$ (the fourth column) was −0.1477 and significant at the 5% confidence level, indicating that the degree of public participation can contribute to the role of the national audit defensive function in the governance of corruption. The results in Table 4 confirm Hypothesis 2: public participation improves the inhibitory effect of national audits on corruption and promotes good governance in the country.

To examine the robustness and validity of the regression findings in Model 1 and Model 2, a new control variable (population (Pop)) was included in the regression Equations (1) and (2).

Table 5. Robustness test of Model 1.

Variables	Preventive Function	Exposed Function	Defensive Function
lnPre	−0.0653 ***	-	-
lnExp	-	−0.0464 *	-
lnDef	-	-	−0.2543 ***
Hr	−0.0108	−0.0106	−0.0175 *
Gov_size	1.5081 *	1.4000 *	1.8772 ***
Wage	−0.0581	−0.0016	−0.0726
Open	−0.3190 **	−0.2241	−0.2289
Gid	−0.1782	−0.1696	−0.1554
Pop	0.0001 ***	0.0002 ***	0.0001 ***
constant	6.6100 ***	6.0320 ***	5.2577 ***
Year	Yes	Yes	Yes
N	300	300	300
${\mathrm{R}}^2$	0.1650	0.1224	0.1529

Note: ***, **, and * indicate that the regression coefficient is significant at the levels of 1%, 5%, and 10%, respectively.

reports the results of the robustness tests for Model 1. The regression coefficients of the national audit function in the second, third, and fourth columns were significantly negative, consistent with the results in Table 3.

Table 6. Robustness test of Model 2.

Variables	Preventive Function	Exposed Function	Defensive Function
${\mathrm{l}\mathrm{n}\mathrm{P}\mathrm{r}\mathrm{e}}_{\mathrm{t}-1}\times \mathrm{T}\mathrm{c}\mathrm{p}$	−0.0740 **
${\mathrm{l}\mathrm{n}\mathrm{E}\mathrm{x}\mathrm{p}}_{\mathrm{t}-1}\times \mathrm{T}\mathrm{c}\mathrm{p}$		−0.0527 **
${\mathrm{l}\mathrm{n}\mathrm{D}\mathrm{e}\mathrm{f}}_{\mathrm{t}-1}\times \mathrm{T}\mathrm{c}\mathrm{p}$			−0.1873 ***
Tcp	−0.2576	−0.6963 **	−1.5039 ***
Hr	−0.0015	−0.0137	−0.0129
Gov_size	1.7768 **	1.6478 **	1.5286 **
Wage	−0.0676	−0.02123	−0.0207
Open	−0.4403 **	−0.4618 ***	−0.4119 **
Gid	−0.0792	−0.1279	−0.0946
Pop	0.0002 ***	0.0002 ***	0.0002 ***
constant	6.7039 ***	6.7194 ***	6.6169 ***
Year	Yes	Yes	Yes
N	300	300	300
$R^2$	0.1439	0.1140	0.1667

Note: *** and ** indicate that the regression coefficient is significant at the levels of 1% and 5% respectively.

reports the results of the robustness tests for Model 2. Similar to the results presented in Table 4, the regression coefficients of the national audit preventive function and national audit defensive function in the second and fourth columns of Table 5 were significantly negative. Thus, Hypothesis 2 is supported by the findings.

5. Discussion

The empirical findings confirmed Hypotheses 1a, 1b, and 1c. This means that the prevention, exposure and defensive functions of national audits allow declining corruption. It should be noted that our empirical results are coherent with past studies [54,55,56,59,60,61]. The “immune system” of the national audit includes preventive, exposed, and defensive functions, which have inhibitory effects on corruption. However, as in the study [54], this investigation showed that the effects are not evenly balanced, and the defensive function has the most-significant deterrent effect. As a crucial component of national governance, the national audit plays an “immune system” role in the process of achieving “good governance”. It forms three lines of defense to prevent, expose, and defend against corruption and prevent corruption from proliferating. The preventive function focuses on advanced supervision, which is the basis of the “immune system” function. Due to the characteristics of the national audit itself, it has a strong deterrent force. The exposed function focuses on supervision during the event, which is the core of the “immune system” function. It exposes various corruption problems that manifest in economic activities, exposes corrupt behavior in the open, and plays a deterrent role in suppressing corruption. The defensive function is similar to post-event supervision, which is the primary focus of the “immune system” function. The audit investigates and handles all kinds of corruption cases, transfers them to judicial authorities, punishes all types of corruption, provides audit recommendations, supervises rectification and implementation, and plays the most-significant deterrent role in curbing corruption. The preventive function of the national audit is more effective in areas with higher public participation in governance, while the exposed function does not show significant effects. This suggests that the national audit’s exposed function is largely driven by its own fulfillment of responsibilities and is less influenced by external factors.

The findings confirm Hypothesis 2 that public participation in the national audit process enhances the audit’s deterrent force and promotes the implementation of corrective actions. In the administration of national audits, the public plays not only an “evaluator” role, but also a “builder” and “watchdog” role. The engagement of public participation mechanisms strengthens the national audit’s “immune system” function, promotes democratic participation, improves decision-making quality and credibility, reduces conflicts of interest, minimizes administrative costs, and ultimately, contributes to achieving the goal of “good governance”. Furthermore, the coefficient of government size (Gov_size) was positive, which is similar to the results of Kotera et al. [92], demonstrating that government size is positively correlated with corruption. The growing scale of the government will lead to an increase in relevant resource elements, such as administration fees, redundant personnel, political resources, economic resources, and approval processes. This will lead to increased rent-seeking behavior, bureaucracy, and corruption [93,94,95]. The coefficients of Wage, Open, and Gid were negative, indicating that civil servants’ salaries, the level of openness, and government transparency are negatively correlated with corruption. This supports the positive impact of “high salary keeps clean” on corruption [96,97], improving the degree of openness to the outside world, which is conducive to reducing corruption [83,98], and the construction of government information openness can inhibit the growth of corruption [99,100,101].

6. Conclusions

Based on the concept of the “governance perspective”, this study analyzed the role of national audits in corruption governance across 30 provinces, municipalities, and autonomous regions in China from 2008 to 2017. The results demonstrated that the national audit contributes to the “immune system” function of combating corruption to a certain extent and enhances the guidance and effectiveness of its governance. Thus, the findings confirmed the research hypothesis that the “immune system” function of national audits helps to prevent corruption within the preventive, exposed, and defensive functions of national audits. In addition, the second hypothesis that public participation in national audits helps to curb corruption and promote good governance was confirmed.

Considering the results, the study provides meaningful implications for policymakers on the national audit supervision system:

First, the “immune system” of national audits should function in a scientific and rational way so that it can allow its preventive, exposed, and defensive functions to play their role in governance corruption. The results of the empirical analysis in this study showed that the influence of these three functions is quite uneven. Therefore, at this crucial point of economic construction and social development, the following functions of national audits should be properly arranged.

Strengthening the preventive function of national audits: To enhance the preventive function of national audits, the audit process should be initiated in a timely manner, and potential risks should be identified early on. This can be achieved by establishing a systematic and scientific early warning mechanism, which would facilitate risk analysis and risk management. Additionally, the auditing process should be integrated into the decision-making process of government entities and public institutions to ensure that audits are conducted in a timely and effective manner.

Development of a national audit supervision system: To improve the modern audit supervision system, the government should develop a comprehensive and integrated national audit supervision system. This system should be designed to facilitate public participation and should provide greater transparency and accountability. To achieve this, the government should establish a national audit supervision center, which would serve as a platform for public participation in audit activities. The center should also provide training and guidance to auditors and develop and implement audit standards and procedures.

Establishment of an independent audit oversight body: To ensure the effective operation of the national economy, an independent audit oversight body should be established. This body would have the responsibility of ensuring the quality and integrity of audit activities and would provide an additional level of assurance to the public that audits are conducted in a fair and unbiased manner. This would help to enhance public trust in the audit process and would promote greater accountability among government entities and public institutions.

Second, the public should be actively encouraged to participate in national audits to improve the modern audit supervision system. To mobilize public enthusiasm to participate in the process of national auditing, the government should develop a comprehensive audit announcement system. This system should provide timely and accurate information about audit activities, including the scope and objectives of audits, audit findings and recommendations, and the actions taken by auditees in response to audit reports. The system should be designed to be user-friendly and should be accessible to the public via a variety of channels, including the Internet, social media, and mobile devices. Furthermore, to ensure the effective operation of the national audit supervision system, the government should invest in capacity building for auditors and audit supervisors. This can be achieved by providing training, resources, and technical assistance to auditors, as well as by establishing a certification system for auditors. Additionally, the government should promote knowledge sharing and exchange among auditors and audit supervisors to facilitate the dissemination of best practices and innovative approaches to audit supervision.

There are some limitations of this study. First, the index of the national audit function needs to be further improved. The enumerations of relevant data in the national audit in different years of the China Audit Yearbook are inconsistent. Therefore, this study sorted out the indicators of existing data based on the literature as the trade-offs of the study. Second, administrative, judicial, and other supervisory factors also play a certain role in corruption governance. Investigating the relationship between the judicial authorities and the national audit will improve the policy of corruption governance. Further research is necessary to extend the time for investigation.