Impact of COVID-19 on audit fees in Chinese family firms: exploring the roles of ownership, management and directorship Open Access

Despite extensive research on family firms and audit fees, critical research gaps persist. Studies have predominantly focused on family firms in North America and Europe with comparatively limited attention paid to emerging economies (Homayoon and Hakimzadeh, 2017). For example, Chinese family firms, shaped by Confucian principles that emphasise family loyalty and hierarchical control (Dou et al., 2020), exhibit governance structures that diverge markedly from Western models, potentially resulting in distinct audit fee dynamics. In this regard, the relationship between family control and audit fees remains inconclusive. Some studies suggest that family involvement alleviates Type I agency problems (i.e. conflicts between owners and managers), lowering audit fees through the alignment of interests (e.g. Khan et al., 2015). Other studies indicate that family dominance increases audit fees by heightening concerns about inadequate oversight or by exacerbating Type II agency problems (i.e. conflicts between majority and minority shareholders) (e.g. Khan and Subramaniam, 2012). Moreover, previous research offers inconsistent evidence on how audit fees respond to external shocks. While Chen et al. (2018) found that fees decreased under uncertain conditions, Alexeyeva and Svanström (2015) documented an increase.

Of particular interest is the COVID-19 pandemic which profoundly disrupted global economies, presenting unprecedented challenges to audit practices (Zhang et al., 2020), leaving family firms to a unique pressure due to their concentrated ownership and distinct governance structures (Ramos et al., 2016). During COVID-19, Al-Okaily (2020) reported higher fees in UK family firms, a trend echoed by Al-Qadasi et al. (2023) in Oman, underscoring the importance of context-specific analysis. Although recent studies by Cheng and Sutunyarak (2023) and Rahman et al. (2023) have advanced the knowledge of audit fees in China, these studies have not fully accounted for Confucian governance or crisis-specific effects.

This study, therefore, aims to investigate how family ownership, management involvement, and directorship roles influenced audit fees in Chinese family firms during the COVID-19 pandemic. Analysing 18,889 firm-year observations for the 2015–2022 period, with robust controls for endogeneity, the research reveals the following key findings: 1) an insignificant relationship between family ownership and audit fees generally, but a significant negative relationship during COVID-19, and 2) significant lower audit fees associated with family involvement in management (but not directorship) during the whole period under study without any significant alteration due to the pandemic.

The findings contribute to the literature in several ways. First, by focusing on Chinese family-owned listed firms during the pandemic, this study extends the work of Cheng and Sutunyarak (2023), who found that economic policy uncertainty increases audit fees in Chinese firms. In contrast, we demonstrate that family ownership significantly reduced audit fees during the pandemic, suggesting that family governance mitigates uncertainty-related risks more effectively than broader internal control mechanisms. Second, it refines the observation by Rahman et al. (2023) that Chinese family firms face less severe Type I risks but heightened Type II conflicts, typically leading to higher audit fees. Our results show that during a crisis, high family ownership tends to reverse this trend, lowering fees as reduced Type I problems outweigh Type II clashes. Third, in contrast to Al-Okaily (2020) and Al-Qadazi et al. (2023), who found increased audit fees in other countries during the pandemic, our study indicates lower audit fees in China, underscoring the moderating influence of Confucian governance and political connections (cf. Xu et al., 2015). Finally, by providing empirical evidence that the balance between Type I and Type II problems shifts toward reduced audit fees in Chinese family firms during a crisis, this study enriches agency theory discussions within non-Western contexts.

The remainder of the paper is structured as follows: Section 2 presents China’s cultural and institutional environment. Section 3 presents the theoretical framework, and hypothesis development. Sections 4 and 5 outline the research design and empirical results, respectively. Section 7 discusses the findings, and section 8 concludes the paper.

China’s unique cultural and institutional environment makes it a good setting for studying family firms, particularly in the context of audit fee dynamics during external crises. Institutionally, political connections (Xu et al., 2015) and China’s stringent COVID-19 response (Hay et al., 2021) introduced audit challenges distinct from other regions (Harjoto and Laksmana, 2022).

At the heart of Chinese family businesses lies Confucian philosophy, which profoundly shapes their governance and operational strategies (Dou et al., 2020). Confucian values – such as familial loyalty, benevolence, and hierarchical order – cultivate a management style where founders, often acting as both owners and chief executives, exert centralised control (Chen et al., 2013; Redding, 1994). This founder-dominated approach emphasises trust in family members, who frequently occupy key managerial roles, over external professionals. In a study of family enterprises in Guangdong province, Lee and Li (2009) found that 84% had family members in top management, with 95% involving both spouses. This reliance stems from several factors: loyalty to mitigate uncertainty in a volatile economic environment, a scarcity of trustworthy professional managers, financial dependence on family networks, and the willingness of relatives to work for lower compensation during tough times (Lee and Li, 2009).

The Confucian influence fosters a paternalistic organisational climate, where founders adopt a benevolent yet authoritative stance, promoting frugality, self-sacrifice, and family welfare (Chen et al., 2013). This management style tends to reduce Type I agency risks by aligning interests within the family but can exacerbate Type II setbacks when resources are diverted for familial benefit, especially during economic downturns (Dou et al., 2020). This trade-off, influenced by China’s cultural and institutional environment, underscores the novelty of our study.

Moreover, the centralised decision-making by founders – often handling both major and minor decisions personally – limits organisational complexity and professionalisation (Redding, 1994). These traits distinguish Chinese family firms from their Western counterparts that often rely on professional managers (Rahman et al., 2023), and amplify governance challenges, such as succession, intensified by the one-child policy (1980–2015), which restricts the pool of potential family successors (Tang et al., 2024).

Institutionally, China’s socialist market economy, evolving since the 1992 economic reforms, intertwines family firm operations with the political landscape. Approximately 41% of founders maintain robust political connections, securing competitive advantages like resource access or regulatory favour (Xu et al., 2015). Political connections can also serve other functions. For example, politically connected family-controlled firms tend to be more involved in social activities than those without such ties (Ananzeh et al., 2023). However, these connections simultaneously heighten governance risks, including reduced accounting transparency and increased fraud potential, as agency problems intersect with an imperfect business environment characterised by weak property rights protection. Combined with the Confucian-influenced founder involvement, these dynamics create a complex backdrop for analysing how family control structures influence audit fees during crises like COVID-19 in China (Tang et al., 2024).

Ownership reforms in Chinese family firms have been shaped by the country’s transition toward a market-oriented economy. This transformation has simultaneously reinforced family control and generated new pressures for transparency and accountability. Since the economic reforms of 1992, family firms have increasingly consolidated ownership structures to preserve control in the face of intensifying market competition and heightened regulatory scrutiny (Xu et al., 2015). Such ownership concentration is consistent with Confucian values emphasising familial loyalty and cohesion. Meanwhile, regulatory initiatives introduced by the China Securities Regulatory Commission (CSRC) have strengthened disclosure requirements and minority shareholder protections, seeking to balance family dominance with external accountability (Jiang and Kim, 2020). Collectively, these reforms have heightened the demand for rigorous auditing practices to ensure regulatory compliance and enhance stakeholder confidence.

Management reforms have sought to reconcile Confucian-inspired family control with the growing need for professionalisation aimed at enhancing competitiveness. As noted by Lee and Li (2009), the traditional reliance on family members for managerial positions is rooted in trust and loyalty, yet such practices may constrain access to professional expertise and heighten governance risks, particularly during periods of volatility (Tang et al., 2024). Recent reforms have promoted the appointment of professional managers to improve operational efficiency and mitigate Type II agency risks, especially within firms characterised by strong political connections (Xu et al., 2015). Nevertheless, these management reforms reveal a persistent tension between the preservation of cultural traditions and the adoption of modern governance principle. This tension may also affect audit practices, necessitating audit approaches that are sensitive to the degree of family involvement (Hay et al., 2021).

Audit reforms in China have substantially reshaped the oversight of family firms, particularly in response to global crises and evolving domestic regulatory frameworks. Since the early 2000s, the CSRC, together with other supervisory authorities, has strengthened audit standards by emphasising auditor independence and audit quality. These reforms have been especially critical in mitigating governance risks within family firms, which often exhibit reduced transparency due to political connections (Xu et al., 2015).

Agency theory rests on the assumption that rational agents (i.e. managers) act primarily in their own self-interest, which may give rise to agency conflicts when their objectives diverge from those of the principals (i.e. owners) (Jensen and Meckling, 1976). In an agency theory framework, financial auditors assess the level of agency conflicts within firms to determine the scope of their efforts, aiming to reduce information asymmetry between principals and agents (Öhman et al., 2006). Firms with elevated agency conflicts require more audit work, which in turn raises audit fees (Schierstedt and Corten, 2021). However, family firms tend to experience distinct agency dynamics compared to non-family firms.

Type I agency problems stem from the separation between owners and managers. This type of problems is particularly relevant in large companies, where owners often have limited insight into how management runs the company, which in turn is linked to large degrees of information asymmetry (Eisenhardt, 1989). In family firms, family members often wield significant control over operations due to ownership concentration and active participation in management. This involvement reduces information asymmetry between owners and managers, as family members directly monitor activities (Chen et al., 2008; Muñoz-Bullón and Sanchez-Bueno, 2011). As a result, the risk of managerial manipulation of financial reports decreases, lowering the likelihood of misstatements (Al-Okaily, 2020). Auditors perceive less need for extensive audit procedures under these conditions, leading to reduced fees. Moreover, the revenue of family firms is typically intertwined with family income, fostering a long-term orientation that discourages short-term financial manipulation (Al-Okaily, 2020). This alignment of interests further mitigates Type I risks, decreasing the demand for audit services.

Type II agency problems arise from conflicts between majority and minority shareholders. Also in this case, conflicts between the two groups of shareholders are evident in large companies. However, theoretical arguments suggest that type II problems may likewise be prevalent in family firms. Family members’ dominance in management and board positions can enable them to prioritise their private interests, potentially at the expense of minority shareholders, especially during crises (Villalonga and Amit, 2006). This necessitates greater audit oversight to safeguard minority shareholders’ rights. Additionally, family involvement in governance may lead to ineffective oversight if family members lack sufficient management expertise (Wang, 2006), increasing auditors’ workload and, consequently, the audit fees. It should also be emphasised that the severity of agency problems to which minority shareholders are exposed varies considerably across countries, which at least partly can be attributed to differences in the legal protection afforded to these shareholders (La Porta et al., 2000).

In Chinese family firms, Confucian values place emphasis on familial loyalty and collective welfare rather than individual gain (Dou et al., 2020). This means that the agency theory assumption of self-interest tends to be less pronounced, as family owners and managers often share aligned interests to ensure the firm’s long-term survival (Lee and Li, 2009). Such alignment helps to reduce Type I agency costs, and thereby also perceived audit risk (Chen et al., 2008). However, the potential for opportunistic behaviour remains a concern, as concentrated ownership structures can intensify Type II agency conflicts. This situation necessitates audits to protect minority shareholders (Rahman et al., 2023), and highlights how family control (i.e. involvement in management and directorship) may influence information asymmetry and the scope of audit procedures (Al-Okaily, 2020).

External shocks are also critically linked to agency risks (Schierstedt and Corten, 2021). During the COVID-19 pandemic, the interplay between Type I and Type II problems in Chinese family firms became evident in three respects. First, the crisis underscored the stability and long-term orientation of family firms, where aligned family interests foster resilience and prudent management (cf. Al-Okaily, 2020). Auditors likely perceived this stability as reducing overall risk, thereby diminishing the need for extensive audit procedures despite the potential for Type II conflicts. Second, Confucian values emphasising familial loyalty reinforced the alignment of interests within these firms (Dou et al., 2020). During the pandemic, such cultural norms may have curtailed the likelihood of family members exploiting minority shareholders, thereby mitigating Type II conflicts and alleviating auditors’ concerns. Third, China’s stringent COVID-19 response, including nationwide lockdowns (Pei et al., 2020), travel restrictions (Zhang et al., 2023), and government support policies (Wang et al., 2021) such as tax relief, low-interest loans, and subsidies for firms, as well as enhanced regulatory oversight (Hay et al., 2021; Wang et al., 2021), helped stabilise firm operations and cash flows. These measures reduced financial pressure on controlling families and diminished their incentives to extract private benefits from minority shareholders (Jiang and Kim, 2020), thereby lowering the risk of expropriation by majority owners. The mitigation of Type II problems, supported by increased liquidity and transparency, decreased the perceived risk of misstatement or fraud, which in turn reduced the need for extensive audit procedures and ultimately decreased audit fees (Hay et al., 2021).

Agency theory further suggests that effective governance reduces information asymmetry and the need for extensive audit efforts (Eisenhardt, 1989). Governance quality is often reflected in characteristics such as the frequency of board meetings, board size, and the proportion of independent directors (Liu and Hu, 2006; Wu, 2012). Moreover, firm-specific factors such as leverage, return on equity, and return on assets capture aspects of financial risk that influence the level of audit effort required. During periods of external shocks, these factors interact with family governance structures, illustrating how Confucian-inspired trust and government can reduce perceived risks (Hay et al., 2021). Finally, engagement with Big 4 auditors typically signals higher audit quality, which may command premium fees (Ho and Kang, 2013). However, in Chinese family firms, strong familial alignment often serve to offset such costs (Khan et al., 2015).

Agency theory posits that audit fees are influenced by the extent of agency conflicts (Eisenhardt, 1989; Jensen and Meckling, 1976). Higher levels of these conflicts typically require more extensive audit efforts. In family firms, concentrated ownership tends to reduce information asymmetry and thereby mitigate Type I agency risks (Chen et al., 2008). However, Type II agency problems may arise when family owners prioritise their private interests over those of minority shareholders. Such a situation may require enhanced audit oversight (Villalonga and Amit, 2006). Moreover, external shocks can exacerbate agency risks, resulting in a demand for increased audit effort.

A key finding in research on audit fees in family firms is that these firms often incur lower audit fees than non-family firms, a pattern commonly explained by family involvement, reduced information asymmetry, and fewer Type I problems (Ben Ali and Lesage, 2014; Ho and Kang, 2013; Khan et al., 2015). Such alignment decreases the risk of financial misreporting and lowering audit efforts (Al-Okaily, 2020). Bolor-Erdene et al. (2024) further noted that local CEOs in family firms lower business and misreporting risks, contributing to reduced audit fees. Conversely, other studies reported that family dominance can increase audit fees due to Type II problems (Khan and Subramaniam, 2012; Wang, 2006). Within the agency theory framework, mixed findings have further been reported in emerging countries such as Malaysia regarding the relationship between managerial ownership and audit fees. Consistent with the theory, Mustapha and Ahmad (2011) found that managerial ownership has an inverse relationship with total monitoring costs, whereas Nelson and Mohamed-Rusdi (2015) reported a non-significant relationship between managerial ownership and audit fees.

The impact of external crises on audit fees introduces further complexity. While Chen et al. (2018, 2019) reported reduced audit fees during the 2008 financial crisis, Alexeyeva and Svanström (2015) found higher fees, attributing this to increased audit risk. During COVID-19, Al-Okaily (2020) and Al-Qadasi et al. (2023) observed higher audit fees in family firms in the UK and Oman, respectively, reflecting greater audit efforts amid economic uncertainty. Harjoto and Laksmana (2022) further noted that lockdown stringency increased audit fees globally, particularly for non-US firms with high audit risk. Previous studies show that Type II risks may intensify during economic downturns like COVID-19. Family owners, leveraging significant voting rights, might prioritise family interests over minority shareholders, increasing expropriation risks (Schierstedt and Corten, 2021; Villalonga and Amit, 2006). This necessitates greater audit oversight, raising fees (Rahman et al., 2023; Salvato and Moores, 2010). Family firms may also hire high-quality auditors (i.e. Big 4-firms) to signal credibility, further increasing costs (Ho and Kang, 2013; Khan et al., 2011).

The Chinese context should also be emphasised. Confucian values promote trust and loyalty, potentially lowering audit fees in normal settings (Chua et al., 2012; Dou et al., 2020). During the pandemic, China’s stringent containment policies (Pei et al., 2020; Zhang et al., 2023), and extensive government support (Hay et al., 2021; Wang et al., 2021), alleviated financial pressures and helped stabilise family firm operations. Although these alignments may mitigate Type I risks, they may simultaneously give rise to Type II conflicts and greater uncertainty during periods of crises. Consequently, the overall impact of family ownership on audit fees in the Chinese context during the COVID-19 period remains ambiguous.

Taken together, agency theory arguments, mixed findings reported in previous studies, and the Chinese cultural and institutional setting lead us to formulate a nondirectional (i.e. null) hypothesis on the relationship between family ownership and audit fees in terms of a crisis. Thus, we propose the following hypotheses:

Family involvement in management and directorship introduces dynamics similar to those associated with family ownership. Under normal conditions, agency theory suggests that family managers, through direct oversight, can reduce Type I setbacks by reducing information asymmetry and, consequently, audit risk (Chen et al., 2008; Chua et al., 2012). Conversely, family dominance can increase audit fees due to Type II problems, particularly when family members prioritise personal interests or lack managerial competence (Khan and Subramaniam, 2012; Wang, 2006). Yet, during crises, weak governance due to family members’ limited expertise (Homayoon and Hakimzadeh, 2017), and board dominance (Hope et al., 2012) may further exacerbate Type II conflicts, thereby requiring more audit work (Ben Ali and Lesage, 2014; Khan and Subramaniam, 2012).

Empirical studies report mixed evidence regarding family involvement in management. In emerging economies like Malaysia, managerial oversight has been found to reduce monitoring costs (Mustapha and Ahmad, 2011), while other studies report no such effect (Nelson and Mohamed-Rusdi, 2015). During the COVID-19 pandemic, Al-Okaily (2020) and Al-Qadasi et al. (2023) observed that audit fees increased for family firms due to heightened audit risk, partly attributable to the challenges of managing firms under such extraordinary conditions.

In the Chinese context, family involvement in management is widespread, largely shaped by Confucian-inspired trust (Lee and Li, 2009). Such involvement can be associated with lower audit fees (Chua et al., 2012). However, family directorships may be less effective due to hierarchical governance structures, which can signal a need for more thorough audits (Rahman et al., 2023). Although the prevalence of family managers with limited professional skills (Tang et al., 2024) may heighten audit risk during crises, China’s cultural and institutional environment gives rise to unique agency dynamics. Confucian values foster trust and loyalty in family members, potentially mitigating Type II concerns (Dou et al., 2020). However, the long-term orientation, strong family cohesion, and government support measures observed during the Chinese COVID-19 period support the assumption of a reduced risk of expropriation.

Also when focusing on management and directorship, agency theory arguments, mixed findings in previous studies, and the Chinese cultural and institutional setting lead us to formulate a null-hypothesis. The following hypothesis is proposed:

This study utilised data obtained from firms listed on the Shanghai and Shenzhen Stock Exchanges in China for the 2015–2022 period to allow for a pre- and post-pandemic comparison (i.e. 2015–2018 versus 2019–2022). All data related to audit fees, as well as family firm ownership and management, were collected from the China Stock Market and Accounting Research Database.

Table 1 outlines the data screening process, which resulted in 18,889 observations. To ensure the precision and reliability of the dataset, the following screening measures were applied: listed companies with incomplete data were excluded, as were companies in the financial and insurance sectors, and those with special treatment status. In addition, the study employed the Winsorization method to mitigate the influence of extreme values by trimming the upper and lower 10% of the data.

For H1, we studied the differences in audit fees between two samples of family and non-family firms. Focusing on family firms, we delved into the influence of family member ownership on audit fees under COVID-19. Following previous research (Al-Okaily, 2020; Clatworthy and Peel, 2007; Zaman et al., 2011), we tested the relationship between family firms and audit fees, as well as family ownership and such fees during the COVID-19 period by using the following regressions:

For H2, we first examined the effect of the proportion of family members in management on audit fees during COVID-19. We then focused on the effect of the proportion of family members on the board of directors on audit fees during the pandemic. Following Al-Okaily (2020), Clatworthy and Peel (2007), and Zaman et al. (2011), we tested the relationship between family involvement in management and directorship and audit fees by using the following regressions:

The dependent variable, audit fees, is the aggregate amount that a firm disburses for auditing services (cf. Al-Okaily, 2020). We utilised the natural logarithm of total audit fees (LNAF) to represent the operational conditions of family firms.

Family firms were identified as those where family members either hold the CEO position, occupy seats on the board, or possess at least 10% of the firm’s equity (cf. Al-Okaily, 2020). A firm that meets any of these criteria was classified as a family firm and designated with a dummy variable (FAMILYFIRM) of 1 (and 0 otherwise). FOWN signifies the proportion of firm equity held by family members, FAMILYMAG denotes the proportion of family members serving in managerial roles, and FAMILYDIR represents the percentage of family members holding positions on the board of directors (Al-Okaily, 2020).

According to Wu (2012), it is suggested that publicly listed firms on the Shanghai Stock Exchange with superior corporate governance needs less audit work, thus incurring lower audit fees compared to other firms. As such, we incorporated governance control variables that pertain to the board of directors. The number of board meetings conducted annually, and the total number of board members served as proxies for board diligence and size, respectively. Liu and Hu’s (2006) analysis of the interplay between agency costs and audit pricing for A-share listed companies from 2001 to 2003 revealed that the ratio of independent directors on the board significantly affects agency costs and audit fees. Therefore, board independence was measured as the proportion of independent directors on the board.

To reinforce our findings and referencing prior literature (Al-Okaily, 2020; Clatworthy and Peel, 2007; Zaman et al., 2011), we also introduced firm-specific control variables, which encompass the institutional investors’ shareholding ratio, leverage, return on equity, return on assets, the age of the firm, and whether a Big 4 firm was hired. Table 2 provides the definition of each variable along with their respective database sources.

Table 3 offers the descriptive statistics for the 18,889 observations and the variables mentioned above. Pertaining to LNAF, we note a diversity in charging standards among different audit firms, leading to considerable disparities in audit fees levied on distinct clients. Table 4 shows the Pairwise correlations matrix. The results show that audit fees have a positive and significant relationship with BODMEET, BODSIZE, FIRMAGE, INVESTPROP, LEV, ROE, and BIG4, but a negative and significant relationship with IDR and INVR. Almost all relationships between other variables have strong significance.

To examine the influence of family control on audit fees, we employed a two-way fixed effect model, following the methodologies of Anderson and Reeb (2003) and Srinidhi et al. (2014). The model incorporates dummy variables for each year and each industry represented in the sample. To assess the impact of the pandemic on the relationship between family control and audit fees, we introduced interaction terms and divided the regression analysis into the entire period (2015–2022), the pre-COVID-19 period (2015–2018), and the post-COVID-19 period (2019–2022), facilitating a comparative analysis.

Tables 5 and 6 present the regression outcomes examining the relationship between family firm status (FAMILYFIRM) and family member ownership (FOWN) on the natural logarithm of audit fees (LNAF), respectively. The models include governance variables (number of board meetings, board size, proportion of independent directors) and firm characteristics. Equation (1)’s results for the entire period, pre-COVID-19, and post-COVID-19 periods are shown in Models 1, 2, and 3, respectively. Model 2 indicates a negative, though insignificant, relationship between FAMILYFIRM and LNAF before the pandemic. However, during the pandemic, the coefficient for FAMILYFIRM becomes positive (0.0261) and significant at the 5% level, suggesting a shift toward higher audit fees in family firms.

When analysing family firms as a group using FOWN, Table 6 reveals contrasting results. Models 4, 5, and 6 show a negative relationship between FOWN and LNAF across the entire period, with the coefficient changing from −0.0007 (insignificant) pre-pandemic to −0.0015 (significant at the 1% level) during COVID-19. This indicates a significant negative correlation between family member ownership and audit fees during the pandemic, contradicting H1. This also contradicts generic agency theory arguments which predicts heightened expropriation risks and increased audit efforts (Al-Okaily, 2020; Al-Qadasi et al., 2023; Ho and Kang, 2013).

Tables 7 and 8 display the regression results on the relationship between family members in managerial roles (FAMILYMAG) and board directorship (FAMILYDIR), respectively, and LNAF. Table 7 (Models 7–9) shows a significant negative correlation between FAMILYMAG and LNAF across the whole period, with all coefficients being significant at the 1% level. These findings contradict H2. Table 8 (Models 10–12) indicates a negative, though insignificant, relationship between FAMILYDIR and LNAF, with no significant change during the pandemic. This is in line with H2.

Although time and industry fixed effects were controlled for in the analysis of the relationship between family control and audit fees, the regression results may still be subject to bias. To enhance the reliability of the findings, we conducted a series of robustness tests.

The audit opinion constitutes the primary output of the audit work (Öhman et al., 2006). Accordingly, we included this variable to reinforce the previous findings from an audit perspective. Hay et al. (2008) further reported that internal control and corporate governance influence audit fees. Therefore, we introduced two additional variables, administrative expenses ratio and internal control weaknesses, which are closely related to internal control and agency costs. The modified models for the robustness tests are as follows:

For H1.

For H2,

Tables 9–12 present the results of the robustness test with additional variables related to audit work and the internal control of the firm. The results are consistent with our previously presented findings and demonstrate the robustness of the main results.

A potential concern arising from our analysis is the risk of endogeneity issues or sample selection bias. To mitigate this, we applied the propensity-score-matching (PSM) method, pairing family firms with non-family firms. This strategy minimises the possibility that the outcomes are dictated by systematic variances in firm characteristics (Srinidhi et al., 2014). The PSM technique pairs the treatment group (i.e. family firms) with a control group (i.e. non-family firms), thereby alleviating issues related to self-selection. In line with Al-Okaily (2020), we employed the nearest neighbours matching technique and the logit model for this purpose, and computed the propensity score using a recognised method that considers all relevant factors:

By using the technique of nearest-neighbour matching (caliper 0.03), the family firm sample was matched with the non-family one (cf. Al-Okaily, 2020). According to Peel (2018), this strategy is preferable since the potential bias reduction is larger the closer the participants are matched. First, family firms (dummy variable = 1) were set as the treatment group and non-family firms (dummy variable = 0) were set as the control group. We then selected nine elements (BODMEET, BODSIZE, IDR, INVESTPROP, LEV, ROE, INVR, FIRMAGE, and BIG4 as paired screening variables. The samples in the treatment group were paired with the control group. After pairing, 25,914 valid sample observations were obtained.

The results of the PSM show a t-stat value of 2.10, implying that after matching, family firms and audit fees are still highly correlated (after excluding the bias due to sample selection).

Table 13 presents the bias and corresponding p-values pre- and post-propensity score matching (PSM). For post-matching, all absolute bias values are under 10 and all t-values fall below 1.96. Additionally, all p-values exceed 0.05. These outcomes verify that the study successfully clears the balance test and corroborates the robustness of our foundational regression findings.

While prior literature has predominantly focused on North American and European family firms, the present findings shed light on an underexplored setting where cultural and institutional factors appear to influence governance and audit practices. Our empirical analysis provides insights into the dynamics of audit fees in Chinese family-owned listed firms during the COVID-19 pandemic. Anchored in agency theory and China’s cultural and institutional context, the findings highlight how ownership structure, management involvement, and Confucian governance interact to influence audit fee outcomes.

First, we document a significantly negative relationship between family ownership and audit fees during the post-COVID-19 period (2019–2022), indicating that higher ownership stakes are associated with lower fees. The finding supports the argument that family ownership mitigates Type I agency problems by aligning the interests of owners and managers (Chen et al., 2008; Muñoz-Bullón and Sanchez-Bueno, 2011), thereby reducing the demand for extensive audit procedures (Ho and Kang, 2013; Khan et al., 2015).

However, this finding diverge from evidence reported in Western contexts, such as the UK (Al-Okaily, 2020), where audit fees increased during the pandemic due to heightened audit risk and additional audit effort. This contrast underscores the role of ownership reforms, and China’s distinct cultural and institutional environment. Confucian values emphasising familial loyalty, long-term orientation, and collective welfare (Dou et al., 2020; Lee and Li, 2009) likely reinforced the alignment of interests, thereby reducing perceived audit risk even during an external shock.

The finding that higher ownership is associated with lower audit fees extends the work of Cheng and Sutunyarak (2023), who reported that economic policy uncertainty increases the fees in Chinese firms. Our results suggest that family ownership effectively counteracts uncertainty-related risks. The findings also refine previous evidence – indicating that Type II conflicts tend to outweigh Type I problems in Chinese family firms, thereby leading to higher audit fees (Rahman et al., 2023) – by revealing that during the COVID-19 crisis, familial alignment and state-led stabilisation measures appear to have reduced expropriation incentives, thereby lowering auditors’ perceived risk.

Second, consistent with theoretical expectations, family involvement in management shows a significant negative association with audit fees across the entire period under study. The lack of differences between the pre- and the post-COVID-19 periods supports the notion that familial participation in operational roles consistently enhance interest alignment and reduce information asymmetry (Chua et al., 2012). This outcome can be explained, at least in part, by management reforms in China. Confucian-influenced, trust-based managerial control fosters organisational stability even during periods of crises (Lee and Li, 2009). Moreover, the persistent negative association between family involvement in management and audit fees underscores the pivotal role of familial operational control. This pattern aligns with agency theory’s emphasis on reducing information asymmetry through direct managerial engagement (Chen et al., 2008; Chua et al., 2012) and corroborates recent evidence that local CEOs tend to reduce misreporting risks (Bolor-Erdene et al., 2024).

In contrast, family involvement in directorship shows no significant relationship with audit fees, suggesting that board-level oversight exerts less influence on mitigating agency costs than managerial control. This may reflect the persistence of hierarchical governance structures and limited board independence in Chinese family firms (Rahman et al., 2023). Notably, the COVID-19 pandemic did not substantially alter any of these relationships, underscoring the resilience of family governance structures. The findings indicate that, even amid external shocks, cultural and institutional continuity enabled Chinese family firms to preserve control and organisational coherence.

The findings also underscore the integrative role of institutional support mechanisms, particularly government subsidies, regulatory oversight, and political connections (Wang et al., 2021; Xu et al., 2015). These mechanisms likely reduced both operational and expropriation risks during the pandemic, supporting the view that China’s hybrid market–state system provides stabilising buffers that, in conjunction with familial governance, help restrain audit fee inflation. The pandemic’s limited impact on the two relationships specified in H2 further illustrates the resilience of familial governance structures. Familial loyalty and collective responsibility appear to diminish both Type I and Type II risks, yielding outcomes opposite to those observed in Western economies, suggesting that governance challenges increase audit fees (Khan and Subramaniam, 2012). China’s strong political connections and stringent COVID-19 policies may also have mitigated expropriation risks typically associated with Type II conflicts, contrasting with general expectations of increased fees during crises (Harjoto and Laksmana, 2022).

The results therefore contribute to a more culturally grounded knowledge of audit governance. Rather than perceiving family control as a risk factor requiring costly external monitoring, auditors in China appear to view it as a stabilising force, consistent with the principles of relational trust and long-term stewardship.

This study investigates how family governance influences audit fees in Chinese family-owned limited firms during the COVID-19 pandemic, drawing on 18,889 firm-year observations from 2015 to 2022. The results show a significant negative relationship between family ownership and audit fees during the crisis, along with a consistently negative effect of family management involvement across the whole period under study. These findings diverge from empirical evidence in Western contexts, where audit fees tend to increase during external shocks due to heightened risks (Al-Okaily, 2020; Al-Qadasi et al., 2023), thereby underscoring the unique role of China’s cultural and institutional context.

Theoretically, these results challenge traditional agency theory arguments (Eisenhardt, 1989; Jensen and Meckling, 1976), including the inherent conflicts between owners and managers as well as between majority and minority shareholders. Instead, Confucian values emphasising loyalty and long-term orientation (Dou et al., 2020). This mitigates agency problems between owners and managers, reducing information asymmetry and audit risk (Muñoz-Bullón and Sanchez-Bueno, 2011). Institutional factors, such as political connections and government support (Wang et al., 2021), have helped to limit incentives for the expropriation of minority shareholders (Jiang and Kim, 2020) and further lowering perceived risks during the pandemic. This point to the need for a culturally sensitive adaptation of agency theory in non-Western settings.

Focusing on Chinese family-owned listed firms during COVID-19, this study makes a theoretical contribution by extending the work of Cheng and Sutunyarak (2023), who found that economic policy uncertainty increases audit fees in Chinese firms. In contrast, the present study demonstrates that family ownership significantly reduced audit fees during the pandemic, suggesting that family governance mitigates uncertainty-related risks more effectively than broader internal control mechanisms. It also refines the Chinese family firm observations by Rahman et al. (2023) by showing that, in times of crisis, high levels of family ownership tend to lower audit fees. By providing empirical evidence that the balance between Type I and Type II problems shifts toward reduced audit fees in Chinese family firms during the pandemic, this study further enriches agency theory discussions in non-Western contexts.

Practically, the findings offer valuable insights. Family firms can capitalise on trust-based management to lower external oversight costs and enhance resilience during crises, as emphasised by Salvato and Moores (2010). Auditors can streamline procedures for Chinese family firms by leveraging their lower risk profiles to reduce costs without compromising audit quality (Hay et al., 2021). Policymakers should consider culturally tailored regulations that acknowledge how Confucian governance and institutional support mitigate agency issues in ways distinct from Western contexts.

Limitations of this study include its focus on China and COVID-19, which constrains the generalisability of the findings. We therefore encourage cross-sectional research, and future studies could investigate other regions and/or post-crisis trends. In fact, cultural settings and regulation environments appear to be important drivers of variations in empirical results. For example, the finding that CEO succession from internal origins tends to increase audit fees in Malaysia (Minanurohman et al., 2025) contrasts with the US-based evidence reported by Brockman et al. (2022). These contradictory findings may serve as a departure point for future research, encompassing different types of firms across diverse geographical contexts.

Unfortunately, we were not able to access data for 2023 and 2024. The data employed were derived from corporate reports/disclosures mandated and published in China, but initially released in Mandarin Chinese. The reports for 2023 and 2024 have not yet been translated or released in formats that allow reliable and systematic integration into our dataset. Nevertheless, we believe that the current period (2015–2022) provides a meaningful and robust empirical foundation for addressing our research question and contributions presented. For future studies, we propose longer study periods (ten years or more) to capture potential shifts and longer-term dynamics. Extending the study period beyond 2022 would enable the analysis of long-term trends, particularly in the context of post-COVID recovery, as highlighted by Hay et al. (2021).

The availability of data sets also constrains the scope of variables that can be examined. Future research could incorporate additional variables, such as the quality of internal controls and the effectiveness of audit committees, into regression models to further increase our knowledge of the factors influencing audit fee levels, both under normal conditions and during periods of external shocks.