Government audit supervision and enterprise mergers and acquisitions

ABSTRACT As an important part of the national political system, government audits play a significant role in supervising and restricting enterprises. Since 2010, the National Audit Office of China has repeatedly disclosed the problems regarding enterprises’ M&A to prevent the loss of state-owned assets. This study takes listed companies controlled by the central enterprises of A-shares from 2008 to 2018 as a sample to investigate the impact of government audit supervision on M&A. We find that government audit supervision reduces the number and scale of M&As. Further research shows that government audits can reduce enterprises’ M&A premiums and affect payment methods. Finally, government audits significantly improve the short- and long-term performance of M&As. These findings show that Chinese government audits positively affect corporate governance, curbing M&As that damage enterprise interests. This study provides micro-level evidence of the governance effect of government audits on the scientific decision-making of enterprises.


Introduction
Corporate M&As are important for companies to participate and cooperate in global industrial divisions.The report to the 20th National Congress of the Communist Party of China emphasised the need to leverage the strengths of China's enormous market, attract global resources and production factors with its strong domestic economy, and amplify the interplay between domestic and international markets and resources.This will position China to improve the level and quality of trade and investment cooperation.One important goal is to promote the internationalisation of the RMB in an orderly manner, be deeply involved in the global industrial division of labour and cooperation, and preserve the diversity and stability of the international economic landscape and economic and trade relations.While accelerating the construction of a strong trading nation, stateowned enterprises, relying on their scale advantages, can participate in international competition.Companies must integrate their resources by expanding their scale and multifaceted functions in international trade to obtain competitive advantages and reduce the possibility of encountering bottlenecks.M&A help companies quickly develop domestic and international markets.Companies can improve their reputation and market share in domestic and international markets through M&As.Therefore, M&As are uniquely significant in helping companies participate in global industrial divisions and cooperation.
China's M&A market is massive, and M&As have gradually become an important means of resource allocation for Chinese enterprises.In 2019, Chinese enterprises completed over 2,700 M&A transactions, with disclosure amounts exceeding USD 240 billion.There were 40 transactions exceeding USD 1 billion and 426 transactions exceeding USD 100 million. 1Intuitively, M&A can bring synergies and economies of scale to enterprises, reduce transaction costs with upstream and downstream enterprises, help enterprises achieve diversified operations, and reduce operating risks, significantly improving their overall performance.However, previous research has found that high-quality M&A can help the acquiring party absorb the knowledge of the acquired party and increase R&D investments, which is important in improving supply quality, enterprise quality, and enterprise value (B.Liu & Lyv, 2018;Prabhu et al., 2005).
However, many problems currently exist in corporate M&As, leading to losses after M&As; these require effective restraint and supervision mechanisms.Due to their significant impact on market structure and economic development, corporate M&As have attracted the attention of regulatory authorities.However, many problems still exist in the process of corporate M&As, leading to huge losses after M&As.Under policy burdens and information asymmetry, assessing the performance of state-owned enterprise executives solely based on the 'profitability' indicator is difficult.The 'enterprise growth rate' is a critical evaluation indicator for enterprise development.Under this pressure, corporate executives initiate M&As to achieve rapid growth (S.Chen et al., 2015) and meet the development requirements of expanding the company's scale.However, M&A behaviour that focuses on speed rather than investment quality can also bring serious legal, financing, and integration risks to companies.Overexpansion does not create enterprise value.It can cause considerable losses to the company.One example is from an announcement by the National Audit Office in 2018: A company controlled by the China National Machinery Industry Corporation did not fully consider the risk and acquired a company, which resulted in a loss of CNY 350 million after the acquisition.In state-owned enterprise M&As, problems such as unreasonable market transaction pricing and the 'M&A performance paradox' (H.Pan et al., 2008;S. Zhou et al., 2017) make it difficult for enterprises to obtain the synergy of M&As and the efficiency of mergers and reorganisation (S.Li et al., 2020) and affect the healthy operation and long-term development of the entire national economy.There is a need for supervision mechanisms that can effectively constrain enterprises to carry out high-quality M&A and prevent resource abuse.This is of great practical significance for realising the healthy development of state-owned enterprises and preventing the loss of state-owned assets.
Government auditing, as an important part of the national political system, is an institutional arrangement that uses power following the law to supervise and constrain power (J.Liu, 2012) and has made positive contributions to maintaining national economic security, promoting the comprehensive deepening of reforms, and promoting the construction of a clean government.At the Fourth Plenary Session of the 19th Central Commission for Discipline Inspection, General Secretary Xi Jinping emphasised the need to improve the party and state supervision systems.Restraint and supervision of power can ensure the implementation of various decision-making arrangements, policy measures, and governance by the party and the State.Government audits play a supervising and constraining role in the M&As of state-owned enterprises by revealing corporate issues, thereby regulating development and becoming an effective measure to improve the competitiveness of the state-owned economy and preserve and increase the value of state-owned assets.The existing research has explored the role of government auditing as an independent supervisory force in terms of earnings quality (S.Chen et al., 2013), operating performance (J.Li et al., 2015;L. Zhang et al., 2015), and anti-corruption (Chu & Fang, 2016;W. Zhou et al., 2017).Related research provides valuable ideas for strengthening anti-corruption governance and improving operating efficiency (Dou et al., 2019;W. Zhou et al., 2017).However, the impact of government audit supervision on M&A behaviour remains to be explored.This study explores this perspective.
Since 2010, the National Audit Office has repeatedly revealed problems in the M&A of central enterprises.For example, considering the 35 audit results announced in 2018, 14 central enterprises were involved in 28 acquisition projects that caused losses totalling CNY 9.1 billion in violation.According to the National Audit Office's announcement, huge losses from M&A are mainly caused by the failure to strictly implement internal collective decision-making procedures, inadequate consideration of risks during acquisition, insufficient consideration of internal organisational risk warnings, insufficient deliberation, failure to follow decision-making procedures, and inadequate post-acquisition operation management.Therefore, after being inspected by the National Audit Office, will M&A decision-making be affected?Do government audits affect the number and scale of M&As?Will it also impact the premium, payment method, and short-and long-term performance of M&As?The answers to these questions have profound theoretical and practical implications.
This article researches the abovementioned issues based on a sample of A-share listed companies controlled bycentral state-owned enterprises in China from 2008 to 2018.This study shows that government audit supervision reduces the number and scale of M&A, indicating that government audit supervision can reduce M&A behaviour.Notably, government audit supervision is based on the requirements of economic and social development, and its purpose is not to reduce M&A behaviour but to assist enterprises in reviewing the legal, financing, and integration risks they face in M&A activities and to improve their M&A decision-making processes through long-term mechanisms for audit rectification.Further research shows that government audit supervision can reduce M&A premiums and significantly impact the payment methods and the short-and long-term performance of enterprise M&A.The results show that government audits have a positive and effective governance effect, helping and supervising enterprises rectify unreasonable aspects of their M&A decision-making processes and making enterprises conduct more scientific and standardised M&As.
This study makes the following contributions.First, it explores ways to suppress ineffective M&As.M&As have attracted significant attention as an important means to achieve economies of scale, expand markets, reduce transaction costs, and reduce operational risk.Previous studies have explored the influence of policy factors, such as executive political promotion, the government environment, development zone hierarchy, and industrial policies, on enterprise M&A decisions (Q.Cai & Chen, 2020, Qiu et al., 2020;S. Chen et al., 2015;Xiao et al., 2018), whereas this study discusses it from the perspective of effective M&A governance policy tools.This study shows that government audits, critical external governance subjects for enterprises, play a role in M&A governance measures, prompting managers to make more scientific and effective decisions.This study provides policy tools to curb ineffective M&A behaviours, which has enlightening significance for improving M&A performance.
Second, it supplements the research on the economic consequences of government audits.Since implementing government audits in central enterprises, they have attracted widespread attention from domestic scholars.The existing literature has discussed the economic consequences of government audits from the perspective of social audit pricing and fees, corporate innovation, and anti-corruption (L.Chen et al., 2016;Guo et al., 2021), but there are few studies from this perspective of managerial decisionmaking.This study explores the role of government audits in promoting scientific decision-making for enterprise M&As from the perspective of M&A decision-making.Further, it investigates the impact of government audits on M&A premiums and performance.This study provides a helpful supplement to the literature.It has practical significance for promoting the continuous role of government audits, establishing and improving a longterm mechanism for rectification, improving the competitiveness of state-owned enterprises, and preserving and increasing the value of state-owned assets.

Corporate mergers and acquisitions
Extensive research has been conducted on the motives for corporate M&As, divided into three main aspects: synergy effects, managerial overconfidence, and agency problems.Among them, the synergy effects motive theory explains how corporate M&As can create value, while the managerial overconfidence and agency problems motive theories explain why M&As can damage a company's value and why losses from M&As are a key concern for government audits (B.Liu & Lyv, 2018;X. Zhang, 2003).
First, the synergy effects motive theory states that enterprises expect to reduce production and transaction costs by expanding their scale, occupying a larger market share and improving their competitiveness (Moeller et al., 2004).Porter (1985) pointed out that 'synergy' is the aggregation between business units of enterprises and includes the interdependence of various businesses.Braguinsky et al. (2015) found that after M&A, enterprises effectively reduce their inventory rates and management costs through internal inventory allocation, planning, and integration.Blonigen and Pierce (2016) believed that enterprises integrate and restructure non-production activities through M&As, reducing the variable costs of non-production activities such as advertising, thereby exerting the 'economies of scale' effect.G. Jiang (2021) believed that a fast and low-cost way to enter the export market is to directly obtain channels and marketing resources to enter overseas markets through M&As, far lower than directly establishing market and marketing networks overseas.
Second, the managerial overconfidence motive theory indicates that M&As are motivated by managers' underestimation of uncertainty, which often increases the number of M&As and leads to overpayment.Overconfidence is a psychological bias in which managers overestimate the operating environment of the company, their abilities, and the value of the target company, thus overestimating the possibility of M&A success.Zhu and Ma (2021) found that overconfident managers often overestimate their ability to control events and solve problems, thereby implementing unreasonable non-efficient investments and increasing the risk of goodwill impairment.Roll (1986) first proposed that overconfident managers are an important reason M&As cannot create value or cause value damage.Managers optimistically believe they can achieve successful M&A performance, so they overinvest when they have abundant internal funds (Malmendier & Tate, 2005).However, in this case, managers still aim to maximise shareholder interests, avoid obtaining more personal benefits, or ignore M&A performance.However, some studies believe that the relationship between managerial overconfidence and M&As is insignificant in China (F.Jiang et al., 2009).
Third, agency problem motive theory indicates that managers expect to pursue their interests through corporate M&As, which often contradicts the goal of maximising enterprise value.Managers are motivated to obtain personal benefits in agency problems by deliberately investing in unreasonable M&A projects (Zhang and Yuan, 2013).Executives desire empire-building (Jensen, 1986), and company size is usually one of the determinants of managerial compensation.Therefore, managers make more frequent M&A decisions due to the desire to increase monetary compensation and implicit perk consumption (Grinstein & Hribar, 2003).Managers are strongly motivated to obtain personal benefits by launching large-scale M&A activities to expand the scale of the enterprise and acquire assets, thereby increasing compensation (Fu et al., 2014).Sun et al. (2021) found that executives of acquiring hometown companies cause listed companies to pay higher premiums, resulting in goodwill impairment, with low efficiency in merger and acquisition integration.

Government audit supervision
We focus on audits of the financial revenue and expenditures of central enterprises conducted by the National Audit Office of the People's Republic of China.The scope of its audit includes financial revenue and expenditure and the problems in managing bidding, foreign investment, and the internal management of enterprises.This supervision and audit, carried out by the National Audit Office, is a random inspection of central enterprises.Every year, different numbers of central enterprises are inspected, and the audit results are published on the official website of the National Audit Office the following year.The intervention period for government audits is usually one month, and their impact on enterprises may be sustained.Previous studies also explored the supervision function of government audits, mainly focusing on corruption, business, and risk governance.
First, government audits play a role in governance by revealing corruption and improving the efficiency of anti-corruption.In contrast to social audits, government audits, as third parties, have stronger independence; corruption and bribery can be curbed by auditing a company's power operation.In their earliest foreign research, Ferraz andFinan (2008, 2011) find that the disclosure of corruption in government audit reports can influence political elections.Subsequently, studies on government audits to curb corruption were conducted in China.J. Liu (2012) find that government audits have a deterrent effect on corruption.L. Chen et al. (2016) find that government auditing can reduce the number of cases of corruption and bribery in provinces in the following year and reduce excess expenditure on road construction projects.W. Zhou et al. (2017) find that government audits increase the probability of exposure to corruption in inefficient investment enterprises.
Second, government audit plays a role in business governance by improving enterprise governance and enhancing enterprise innovation ability.It enables enterprises to achieve better economic benefits, assume better responsibilities for people's livelihoods, promote the deepening reform of state-owned enterprises, and safeguard the value added of state-owned capital.L. Zhang et al. (2015) found that government audits can improve the business performance of enterprises by improving their resource allocation efficiency.J. Tang and Cai (2021) found that government audits could improve capacity utilisation rates by supervising policy implementation.Dong and Zhang (2021) showed that the stronger the government audit, the more beneficial it is to promote the high-quality development of enterprises.
Third, government audits play a role in risk governance by promptly disclosing negative information about enterprises.The National Audit Office has repeatedly emphasised its goal of preventing and defusing significant risks, expecting to help enterprises defuse risks and difficulties in various ways.Chu and Fang (2017) researched the impact of government audits on stock price crash risk and found that government audit intervention mitigated enterprise risk.L. Cai and Zhou (2016) studied listed banks and found that government audits could prevent systematic risk in the banking sector.

Theoretical analysis and hypothetical development
The General Office of the CPC Central Committee and the General Office of the State Council jointly issued the 'Suggestions on Strengthening the Auditing Supervision of State-owned Enterprises and State-owned Capital' in March 2017, which states that 'audit institutions must follow state-owned enterprises and state-owned capital, leaving no blind area'.Therefore, mergers and reorganisation are key matters of government audits as an important measure of the mixed-ownership reform of state-owned enterprises.
In announcing the audit results of central enterprise group companies, the National Audit Office repeatedly disclosed that the central enterprises had not fully demonstrated and failed to follow decision-making procedures for share acquisition, which led to unreasonable pricing in market transactions, severe loss of state-owned assets, low efficiency of enterprise mergers and reorganisation, and other problems.In 35 announcements of audit results in 2018, 14 central enterprises had violated CNY 9.1 billion in lossmaking acquisitions.Government audit supervision may play an important role in avoiding the serious negative impact of such behaviour on the development of government audit supervision.
From the perspective of the synergy motive of M&As, government audit supervision strengthens the synergy motive of M&As (X.Zhang, 2003).Government audits focus on the impact of oil prices, trade friction, tax policies, and other external business environments on enterprises, hoping to help them solve problems and improve their businesses.The National Audit Office announced in a report, 'Neither stop at performance, nor relax to promote reform'.The fundamental goal of government audits is to promote highquality economic development and deepen reform.This goal will influence enterprises' M&A decisions, especially by strengthening the synergy motive for M&A.After undergoing government audits, enterprises can better select M&A projects after carefully investigating the target enterprises in advance and demonstrating the necessity and feasibility of implementing M&A projects that could create value for enterprises.On the one hand, if it is found that there are more potential business risks in M&As during demonstration and decision-making, the enterprise will reduce high-risk and unnecessary M&As.On the other hand, if the M&A project can achieve a good synergy effect, the enterprise will actively promote it.
From the perspective of managerial overconfidence motives in M&As, government audit supervision pays attention to and helps improve the decision-making process of enterprises to restrain managerial overconfidence and reduce M&As.Overconfident managers will optimistically overestimate their capabilities and business environment (H.Pan et al., 2006), overestimate future business performance, and underestimate business risk (Landier & Thesmar, 2008) in investment judgement, thus implementing M&A s more frequently than non-overconfident executives (C.H. Tang et al., 2020), ultimately leading to the loss of enterprise value (Shi & Zhu, 2010).In the audits of enterprise groups and their subsidiaries over the years, the National Audit Office has focused on the problem of huge losses caused by non-compliant behaviours in mergers and acquisitions.The government audit has repeatedly disclosed problems related to M&A decisions in the report, such as 'insufficient risk consideration during the acquisition', 'insufficient consideration of internal institutional risk warning', 'not strictly implementing internal collective decision-making procedures', 'without sufficient demonstration', 'not under decision-making procedures', and 'post-acquisition operation management is not in place'.In addition, government audits focus on the M&A decision-making process to help and supervise the rectification of unreasonable aspects, making enterprises more objective in estimating the risks in the M&A and reorganisation processes, thus providing enterprise managers with a reference for scientific decision-making.Therefore, government audit supervision enables managers to consider risks more carefully and reduce M&A behaviour caused by managers' overconfidence.
From the perspective of the agency motive for M&As, government audits help improve enterprises' internal control, weaken managers' interests, and supervise the operation of power, thus inhibiting agency motives and reducing M&As (S.Chen et al., 2013;Liang et al., 2020).Since manager compensation is usually related to the scale of a company (Grinstein & Hribar, 2003), executives always have a strong desire for 'empire building', driven by compensation benefits (Jensen, 1986).By launching large-scale M&A, executives can expand the scale of their enterprises and acquire assets to increase compensation (Fu et al., 2014).However, this compensation has no significant relationship with post-merger performance (Harford & Li, 2007), so managers may not be punished for M&A failure.In particular, enterprises are more likely to initiate M&A when executives face greater endogenous growth pressure but have higher political promotion opportunities (S.Chen et al., 2015) or when capital market pricing is unreasonable, leading to arbitrage opportunities for companies or individuals (S.Li et al., 2020).The performance of such M&As generally has no significant difference in the short term but is poor in the long term (S.Chen et al., 2015;Tian et al., 2013).Compared with managerial overconfidence motives, driven by agency motives, even if managers understand the risk of the merger decision, they may continue to launch M&A s and damage the company's value because their interest function differs from the enterprise value function.Enterprises must inhibit managers' self-interest pursuing and reduce opportunistic behaviour by strengthening internal controls.From the perspective of the cost-benefit framework of individual managers, launching M&As can seek personal interests, while disclosing problems in M&As by government audits may damage their reputation.Therefore, government audits allow managers to treat M&A decisions more prudently.Government audits are characterised by strong comprehensiveness and high supervision levels.As a subsystem of the national governance system, it audits whether the company's main leaders are implementing the power structure and operation mechanism, established under reasonable structure, scientific configuration, and strict procedures.However, government audits focus on major risks and hidden dangers in the internal management of enterprises and can weaken managers' interests by playing the role of supervision and governance (Chi et al., 2019;Chu & Fang, 2017).In addition, serious and repeated losses from M&A s may affect the political future and reputation of managers; they reduce such problems during government audits.Therefore, government audit supervision can restrain managers' opportunistic and shortsighted M&As and reduce subjective damage to corporate interests (X.Tang et al., 2012).
Based on the above theoretical analysis, we propose a research hypothesis: Hypothesis 1: After inspection by government audit, enterprises will reduce mergers and acquisitions.

Model construction and variable definition
We establish a difference-in-differences model controlling for firm and annual fixed effects to test the influence of government audit supervision on the M&A of listed companies: MA represents M&A behaviour, GA represents government audits, Control represents the control variable of the model, γ t represents the time-fixed effect, and μ i represents the firm-fixed effect of the company.
The main variables involved in Model (1) are defined as follows:

Dependent variables
Following S. Chen et al. (2015) and S. Li et al. (2019), we constructed the dependent variables MA_N and MA_Size.MA_N is the total number of M&As conducted by a listed company in a year.Following S. Li et al. (2020), multiple M&As that occur on the same day can be counted as one.MA_Size is the natural logarithm of the average amount paid by a listed company for an M&A in that year.

Independent variables
We construct a dummy variable, GA, to capture the status of a firm affected by a government audit.GA equals 1 if in the year and the following years when the company is inspected by a government audit, and 0 otherwise.

Control variables
As in F. Jiang et al. (2008) Natural logarithm of years a company has been listed 2 For example, in December 2015, China National Pharmaceutical Corporation proposed to issue shares to China National Pharmaceutical Holdings to purchase its 96% stake in China National Pharmaceutical Beijing.We exclude related M&A transactions to alleviate the concerns that state-owned enterprises M&A transactions is out of strategic and social policy purposes.2020), S. Chen et al. (2015) and Fu et al. (2014).Except that the correlation coefficient of Size and Pb is −0.529, the absolute value of the correlation coefficient between variables is all within 0.5 (untabulated).Meanwhile, the variance inflation factors (VIFs) of the variables in the model regression are all lower than 3, indicating that multicollinearity, which may affect the empirical results, is not a serious problem.

Test results of government audit and merger and acquisition behavior
Table 3 reports the test results of Hypotheses 1.In Columns ( 1) and ( 2), the dependent variable is MA_N.Column (1) reports the result without control variables, and Column ( 2) with control variables.The coefficients of GA are significantly negative at the 5% level.In Columns ( 3) and ( 4), the dependent variable is MA_Size.Column (3) reports the result without control variables, and the coefficient of GA is significantly negative at the 10% level.Column (4) reports the result with control variables, and the coefficient of GA is significantly negative at the 5% level.Regarding control variables, the coefficients of Size are significantly positive at 5% and 1% levels, indicating that the larger the enterprise size, the more inclined to M&A expansion.The coefficient of Rind is significantly negative at the 10% level, indicating that the higher the ratio of independent directors is, the more reluctant firms are to adopt M&A expansion.The coefficients of Cash, Lev, Pb, Growth, Scale, Dual and Top1 are insignificant.The coefficient symbols and significance of the above control variables are consistent with F. Jiang et al. (2008).The results suggest that government audits can reduce the number and the average size of M&As, supporting Hypothesis 1.

Parallel trend test
Model (2) was constructed to test whether the treated and control groups met the parallel trend assumption.Following Luo et al. (2015), variables pre_4, pre_3, pre_2 and pre_1 are constructed to capture whether the treated group firm is in the fourth, third, second, and first year before the government audit inspection.Current is constructed to capture whether the treated group firm is in the year of the government audit inspection.The variables post_1, post_2 and post_3 are constructed to capture whether the treated group firm is in the first, second, or third year after the government audit inspection.If the coefficients of pre_3, pre_2 and pre_1 are insignificant, the treated and control group firms meet the parallel trend assumption, indicating no significant difference in the changing trend of M&A behaviour before government audit inspection.Figure 1 shows the coefficients of the interaction terms of Model (2) and their 95% confidence intervals.The dependent variable in Figure 1 (1) is MA_N, and the dependent variable in Figure 1 (2) is MA_Size.In Figures 1 (1) and 1 (2), the confidence intervals of pre_3, pre_2 and pre_1 are close to zero, suggesting that the coefficients of pre_3, pre_2 and pre_1 are insignificant.This indicates no significant difference in the changing trend of M&A behaviour between the treated and control group firms before the government audit inspection.

Propensity score matching test
We used a propensity score matching test to eliminate selection bias between the treated and control groups.Following Q. Li and Ma (2017) and B. Wang et al. (2017), Size, Roa, Pb, Roc, Lev, Age, Top1, Scale, Rind, Dual, and Pay were used in the one-to-one non-put-back matching.One thousand six hundred thirty-nine firm-year observations were included in the sample.Table 4 presents the results of the propensity score matching test.The coefficients of GA are all significant at the 5% level, consistent with the main result.

Placebo test
To verify that the test results of Hypothesis 1 are not driven by random factors, a placebo test is conducted by constructing variables GA_Fake2, GA_Fake3 and GA_Fake4 to advance the time of government audits by two, three, and four years, respectively.If the coefficients of GA_Fake2, GA_Fake3 and GA_Fake4 are not significant, the results of Hypothesis 1 are not driven by random factors.Table 5 presents the results of the placebo tests.The dependent variable in Columns ( 1)-( 3) is MA_N and the dependent variable in columns ( 4)-( 6) is MA_Size.In Columns ( 1)-( 6), the coefficients of GA_Fake2, GA_Fake3 and GA_Fake4 are insignificant, indicating that the main test results are less likely to be driven by random factors.Additionally, by randomly assigning the inspection years of the government audit for each firm, randomGA constructed to replace GA in the regression.The process was repeated 1000 times, and the t-value distribution of the coefficient of randomGA was drawn.Figure 2 shows that in 1000 regressions with MA_N and MA_Size as dependent variables, the t-value of the coefficient of randomGA is near zero.This indicates that the t-value of the coefficient of randomGA in 1000 placebo tests is not significant overall; therefore, the main test results are not driven by random factors.

Other measures of the dependent variable
Table 6 reports the robustness test results using other measures of the dependent variable, referring to F. Jiang et al. (2008), S. Wang and Dong (2020) and S. Li et al. (2020).We use three measures to test robustness: the ratio of the average payment amount of M&A in the current year to total assets at the beginning of the year (MA_SizeR), the natural logarithm of total payment amount of M&A in the current year (MA_Sizesum) and whether M&A transaction occurs in the firm in the current year (MA_Dummy).In Column (1), the coefficient of GA is significantly negative at the 10% level.In Column (2), it is significantly negative at the 5% level; in Column (3), the coefficients of GA are all significantly negative at the 5% level.The results of other measures of M&A behaviour are consistent with the main results, indicating that the main results are not affected by specific measures.

Mechanism test
The purpose of government audits is not to reduce M&As but to assist firms in investigating the legal, financing, and integration risks they face in M&As, thereby promoting corporate development.In this section, we examine the possible mechanisms by which government audits play a role in M&A.First, the types of M&A transactions of listed companies are horizontal, vertical, and hybrid.We construct CrossMaRate and CrossMaNum to measure the synergy motive of M&As.CrossMaRate is the ratio of hybrid M&As to total M&As in the current year.CrossMaNum refers to the number of mixed M&As in a company that year.Second, we construct PositiveForecast and EGap to measure manager overconfidence.PositiveForecast is the ratio of the total compensation of the top three managers to that of all managers.EGap is the ratio of a company's executive pay to its peer averages in the same industry.Finally, we construct Mfee and Overpay to measure the agency problem motive for M&A.Mfee is the ratio of administrative expenses to operating revenue.Overpay is excess compensation for the top three managers.Table 7 reports the results.In Panel A, the coefficients of GA are significantly negative at the 5% level; in Panel B, the coefficients of GA are not significant; in Panel C, the coefficients are significantly negative at the 5% and 1% levels, respectively.This indicates that government audits can strengthen the synergy motive and restrain the agency problem motive of M&As, thus reducing the M&A transactions.

Cross-sectional test
Government audits have different effects on the M&A decisions of firms in different environments.We explore the differences in the impact of government audits on firms with different risk levels based on legal risk and financing constraints.Following Xia and Chen (2007) and Ju et al. (2013), we use the intermediary organisation development, legal scores, and SA index to measure the legal risk and financing constraints faced by firms and divide the observations using the annual median.Table 8 reports the results of the cross-sectional test.Panel A reports the results of legal risk; the coefficients of GA are insignificant in the samples with low legal risk in Columns (1) and (3), while in the samples with high legal risk in Columns (2) and (4), the coefficients of GA are significantly negative at the 5% level, indicating that government audit is more significant in reducing corporate M&A behaviour in firms with higher legal risk.Panel B reports the results of financing constraints; the coefficients of GA are insignificant in the samples with low financing constraints in Columns (1) and (3), while in those with high financing constraints in Columns ( 2) and ( 4), the coefficients of GA are significantly negative at the 5% and 1% levels, indicating that in firms with higher financing constraints, a government audit has a more significant reduction in corporate M&As.These results suggest that the effect of government audits on the reduction of corporate M&A behaviour is more significant among firms facing higher risks and that they assist firms in estimating the environment in which they operate and the risks they face.

M&A premiums and M&A payment methods
Following de La Bruslerie (2012) and S. Chen and Li (2016), we decompose enterprises' M&A behaviour into M&A premiums and payment methods for further study.On the one hand, paying a higher M&A premium is an important factor by which M&As fail to create value for firms (S.Chen et al., 2015).After a government audit focuses on the enterprise's M&A benefits, the acquiring firm will require the target firm to improve accounting information quality.This makes the acquiring firm's shareholders and outside directors estimate the M&A synergies and the M&A price paid more accurately to create value for the firm.On the other hand, de La Bruslerie (2012) argued that in a transaction environment with asymmetric information, acquirers use a percentage of shares in addition to cash payments when they are concerned about the overvaluation of acquired assets, and Ge (2015) found that the M&A performance of the cash payment method is significantly lower than the combination of asset, cash, and debt payments.If a government audit can influence firms' M&A payment methods, it is consistent with the hypothesis of this study, and this exploration helps further consolidate the study's logic.
The 132 M&A transactions of the sample firms in the sample interval were considered for this study 4 The variable M&A premium (Premium) is calculated as (the total transaction price, net assets of the subject of the transaction) net assets of the subject of the transaction.The variable M&A payment method (Paytype) equals 1 if the cash payment method is used in the M&A transaction process and 0 otherwise.5Following A. Pan et al. (2018), Size, Lev, OCF, Roe, Growth, Rind, Scale, Price and RelaSize were used as the control variables.Table 8 reports government audit results and M&A premiums, and M&A payment methods.In Column (1), the coefficient of GA is significantly negative at the 1% level, indicating that government audits reduce M&A premiums.In Column (2), it is significantly negative at the 10% level, indicating that government audits can influence the M&A payment method.The decrease in M&A premiums and the change in M&A payment method indicate that government audits can reduce the M&A behaviour of firms and assist in improving the quality of M&As to promote the quality and efficiency of central enterprises.

Market performance
Government audits make firms more cautious with M&A decisions and improve their information needs about the target company.Knowledge about the target company helps firms improve efficiency in the M&A integration process, making them more likely to realise advantages such as synergistic effects from M&As (Y.Wang & Li, 2017;Rossi & Volpin, 2004). 6Regarding short-run M&A performance, accounting information quality significantly impacts improving M&A performance (H.Pan & Yu, 2014), and M&A transactions that occur after government audits should produce better market responses than those before the government audit.Previous studies have found that establishing national-level development zones inhibits firms' M&A expansion behaviour and improves long-term M&A performance (Q.Cai & Chen, 2020).Then, in terms of long-run M&A performance, the reduction in inefficient M&A transaction behaviour should lead to a corresponding improvement in the long-run M&A performance.Therefore, while government audits reduce the number and size of M&As, they should also improve the shortand long-term performance of firms' M&As. 7revious studies usually measure M&A short-run market performance in terms of cumulative excess return (CAR), regarding Y. Wang and Li (2017) and Ying and He (2020), setting the short-term market performance variables CAR [−1, 1] , CAR [−2, 2] and CAR [−5, 5] for excess return in the windows [−1, 1], [−2, 2], and [−5, 5].Following Zhai et al. (2011) and Tian et al. (2013), Size, Lev, Price, Cash, Pb, RelaSize are the control variables.Referring to H. Pan and Yu (2014) and S. Chen et al. (2015), the long-run performance variables △ROA t-1, t + 1 , △ROA t-1, t + 2 and △ROA t-1, t + 3 are constructed as the changes in the ROA one year, two years, and three years after the first announcement date of the M&A and the ROA one year before the first announcement date of the M&A.Following S. Li et al. (2020) and S. Chen et al. (2015), Size, Lev, Paytype, OCF, Roe, and Price were the control variables.
Panel A in Table 9 reports the results of government audits and short-run market performances.In Column (1), the coefficient of GA is significantly positive at the 1% level;8 in Column (2), it is significantly positive at the 1% level; and in Column (3), it is significantly positive at the 5% level.The results indicate that government audits positively affect governance on corporate M&A transactions by auditing the decision-making process and effectiveness of M&A projects, significantly improving the short-term market performance of corporate M&As.Panel B in Table 9 reports the government audit results and long-run market performance.In Column (1), the coefficient of GA is significantly positive at the 5% level; in Column (2), it is significantly positive at the 5% level; and in Column (3), it is significantly positive at the 10% level 9These results suggest that government audits significantly improve the long-run market performance of corporate M&As.Therefore, the M&A transactions after government audit inspection are more likely to bring advantages enabling firms to improve their efficiency in long-term operations.

Conclusions and implications
We investigate whether government audits affect corporate M&A behaviour.Government audits inhibit the number of M&As and reduce their scale.Further studies find that government audit intervention reduces firms' M&A premiums, significantly influences firms' M&A payment methods, and significantly enhances the short-and long-run performance of firms' M&As.The results remained robust after a parallel trend test, propensity score matching test, placebo test, and other measures of the dependent variable.
This study offers practical insights exploring government audits and corporate M&A behaviour.The government must continuously improve the quality of government audits by improving the accountability by strengthening the link between the audit results and the following year's financial allocation.Simultaneously, government audits should continuously update its audit tools.By establishing a diversified access channel and sharing platform for audit information, the government can better realise the value preservation and appreciation of state-owned assets.

Figure 2 .
Figure 2. Placebo test: the distribution of t value.
CONTACT Nan Lin linnan@fzfu.edu.cnSchool of Financial Innovation Development Research, Fuzhou University of International Studies and Trade, 706, Block B, New Zhongguan Building, Haidian District, Beijing, China Paper accepted by Guliang Tang.

Table 1 .
Definition of main variables.M&As conducted by the company in the current year MA_Size Natural logarithm of the average amount paid by a listed company for M&As in the year GA Dummy variable that equals 1 if in the year and the following years when the company is inspected by government audit and 0 otherwise Size Natural logarithm of total assets at the end of the year Li et al. (2020)5)015), and S.Li et al. (2020), we select Size, Lev, Cash, Pb, Growth, Rind, Scale, Dual, Top1, and Age as control variables.Table1reports the variable definitions and descriptions in the Model (1).
Table 2 reports the descriptive statistics of the variables.The mean of MA_N is 0.031, indicating that the sample companies have an average of 0.031 M&As per year for 2008-2018.The mean of MA_ Size is 0.465, indicating that the average annual amount of M&A business involved in the sample companies during 2008-20183 3 is CNY 18 million.The mean of GA is 0.389, indicating that 38.9% of the observations had been inspected by the government audit.In addition, the statistical values of variables are consistent with S. Li et al. (

Table 3 .
Government audit and mergers and acquisitions.

Table 6 .
Other measures of the dependent variable.