Governmental reforms and earnings management: examining their influence during a crisis Open Access

The increased complexity of today’s economic, environmental and social systems can lead to crises, uncertainty, and risk becoming common issues worldwide (Guaita Martínez et al., 2024). The COVID-19 outbreak was a never experienced and unexpected emergency that has exacerbated the complexity at all levels. Indeed, from March 2020 to May 2023, when that health emergency was officially declared over, the pandemic crisis produced negative effects for not only the whole economy, but also for business (Goodell, 2020; Hu and Zhang, 2021) and the quality of reporting (Hsu and Yang, 2022; Chen et al., 2023). In response to these negative effects, several countries have tried implementing different policies to support enterprises. These include a combination of regulatory, monetary, and fiscal policies to mitigate the negative influence of COVID-19 (Zou et al., 2024). Such policies have varied significantly and led to different outcomes (Tarkom, 2022; Iyer and Simkins, 2022). For instance, in the European Union, emergency countermeasures (such as lockdowns) led those governments (Piva and Guerini, 2023; Fasano et al., 2022) to introduce, amongst others, policies to allow companies to maintain the “going concern” assumption. This paper investigates the potential effects that these policies might produce on the management decision to practice accrual accounting earnings management (EM) to meet certain earnings targets (Schipper, 1989; Larcker and Richardson, 2004; Dechow et al., 2010; Beuselinck et al., 2019).

The motivation for this paper is framed using both institutional and agency theories. According to institutional theory, many types of regulation enable and empower the social actors — especially their actions, practices, and incentives (Ratu and Rahajeng, 2024) — by conferring licenses, special powers, and benefits, despite the visions of repression and constraint conjured up by the very concept of regulation (Scott, 2013, p. 61). Thus, this paper goes beyond agency theory, the most utilized theoretical lens used by scholars involved in the EM field (Naz et al., 2024). EM’s opportunistic view suggests that managers depart from normal business activities with the desire to misinform financial statements (Habib et al., 2022), and theorizes that — in addition to those internal corporate governance mechanisms, which have been extensively investigated in the literature — less severe governmental policies can indeed reduce agency costs (Jensen, 2005) and, in turn, EM.

This paper also examines the effects on EM of one of these policies, that of the Italian legislation that allows companies to avoid having to recapitalize if their share capital falls below the statutory limit due to losses. This policy, as stated by Article 6, § 1 of Decree-Law (DL) No. 23 of 8 April 2020, suspended several articles of the Italian Civil Code. It was extended by DL No. 198 of 29 December 2022, with effects being evident on annual reports issued between 2020 and 2022. The policy objective was to minimize the pandemic’s negative effects on business through, for example, the Italian policy to suspend depreciation (Mattei et al., 2023).

Based on the literature, during the pandemic, EM increased (da Silva Flores et al., 2023; Garfatta et al., 2023; Taylor et al., 2023; Ryu and Chae, 2022; Rahman et al., 2023; Yan et al., 2022), raising the likelihood of being able to overcome the negative effects produced by the pandemic. During the pandemic, by filling a gap in the literature, we expected to find decreased EM after the introduction of such specific Italian legislation. According to our reasoning, considering the strict correlation between the amount of equity and reported earnings, introducing a governmental measure, such as the suspension of recapitalization obligation to alleviate the severity of possible capital erosion in a crisis, should reduce EM practices. However, such practices are no longer enough to guarantee that firms survive after the introduction of such regulation.

To test this hypothesis, we adopted Larcker and Richardson’s (2004) model. To test the robustness of our findings, we used ratio analysis and other regression models. The sample consisted of 14,404 unlisted Italian firms whose accounting amounts for 2018–2021 were downloaded from the AIDA database, where 2018–2019 is considered to be antecedent to the suspension of recapitalization requirements and 2020–2021 is considered to be after such suspension. The data from the regression models confirmed a decrease in the management decision to manipulate earnings after the introduction of the recapitalization suspension when compared to the pre-introduction period.

This paper contributes to the existing knowledge in at least two ways. First, when focusing on a pandemic setting, few papers, if any, have investigated the effect produced by a regulatory measure, such as the suspension of the recapitalization obligation in a developed country like Italy, on the management decisions to manipulate the reported earnings. A regulation’s ability to affect EM (interpreted as management’s action to misrepresent financial information) is not typical of emerging countries, where corporate governance mechanisms are less effective in obstructing EM than in developed countries (Almarayeh et al., 2022; Bao and Lewellyn, 2017). The isomorphism (DiMaggio and Powell, 1983), a fundamental concept of new institutional theory (Rudko et al., 2024), allowed this paper to extend the results achieved in emerging countries to the developed countries, thereby contributing to the literature and providing evidence that does not accept Alvesson and Spicer’s (2018) conclusion that “neo-institutionalism is now facing a ‘mid-life crisis’” or that, “while remaining one of the dominant paradigms of organizational studies, it can no longer produce insightful and novel conceptualizations” (Rudko et al., 2024).

Secondly, although academics have investigated monitoring activity’s ability to reduce agency costs to align shareholders’ and managers’ interests and, in turn, avoid EM behaviours (Ghaleb et al., 2020) while focusing on corporate governance mechanisms (Gerged et al., 2023), this paper provides new insights that less severe governmental policies also matter, as they are able to reduce agency costs and, in turn, EM behaviour. To the best of our knowledge, the only paper that has yet investigated the relationships between EM and governmental policy is that by Ratu and Rahajeng (2024), but it focuses on the ability of the female presence in audit committee to affect EM in the presence of anti-corruption disclosure policies before the 2018 financial crisis.

In light of these considerations, it is not surprising that our findings contrast with the literature that indicates an increase in opportunistic behaviours during the pandemic (da Silva Flores et al., 2023; Taylor et al., 2023; Yan et al., 2022). To this extent, our results have implications for actual practice by showing that accounting quality (Barth et al., 2008) depends not only on accounting standards, but also governmental measures, such as the one investigated in this paper that affected Italian firms during the COVID-19 pandemic. To understand how, it is necessary to consider that less severe governmental policies on recapitalization obligation should avoid underestimating loan loss provisions and/or depreciations, thereby leading to a more conservative accounting. Conservatism is a dimension of accounting quality and plays an informational role, whereby the timely reporting of bad earnings news reduces the information asymmetry (Ball and Shivakumar, 2005) between debtholders and the firms, thereby facilitating access to capital and debt renegotiations (Biddle et al., 2022). In addition to conservatism, the same policies should increase earnings quality thanks to a reduction of opportunistic EM practices that are intended to inflate earnings and book value, which the literature has demonstrated to be very common in periods of crisis like the COVID-19 pandemic.

The remainder of this paper is thus organized as follows. Section 1 presents the context of the research and its relevance with a summary of the main findings, their contributions, and implications for theory and practice. Section 2 contains the related research and the hypothesis development. Section 3 presents the research design. Section 4 includes the sample selection and descriptive statistics. Section 5 summarizes the results. Finally, Section 6 presents the conclusions, limitations, and future developments for continued research.

According to institutional theory (Meyer and Rowan, 1977), institutions are comprised of regulative, normative, and cultural-cognitive elements that, together with their associated activities and resources, provide stability and meaning to social life (Scott, 2013, p. 56). Through regulation, institutions not only provide prohibitions and constraints on the action being taken, but also support and empower activities and the actors (Scott, 2013, p. 58). Indeed, many types of regulation enable and empower social actors and their action, conferring licenses, special powers, and benefits despite the concept of regulation only conjuring up visions of repression and constraint (Scott, 2013, p. 58). From a theoretical perspective, this study is interested in the regulative elements and activities that provide stability to each organization. Thus, the suspension of the recapitalization requirement for Italian companies, the particular focus of this paper was introduced to guarantee stability to each organization throughout the COVID-19 pandemic. As institutional theory suggests, the introduction of such a regulation should have provided incentives to do or not to do certain activities. In particular, it can be explained through institutional theory that variations in institutional characteristics—such as values, norms, and solid governance monitoring—can be a source of pressure that can also affect managers’ incentives (Ratu and Rahajeng, 2024, p. 350) and in particular EM.

This paper considers EM practices, the management actions taken to misrepresent financial information, that the literature frames within agency theory. Such theory, already extensively adopted by academics (Naz et al., 2024), hypothesises the conflict of interest between managers and shareholders (Jensen and Meckling, 1976) that can lead the former to behave opportunistically against the latter. EM practices are a response to such conflict, and its magnitude, according to our arguments that are framed within the institutional theory, might be affected positively by the issuance of a less severe regulation such as the one this paper investigates.

The managerial actions to manipulate earnings can involve accrual accounting and/or real EM (Gaio et al., 2022), thereby producing “accrual accounting EM” and “real activity EM”. The former refers to the manipulation of accruals (Cohen and Zarowin, 2010; Trueman and Titman, 1988; Zang, 2012) whereas the latter refers to the manipulation of real activities (Einhorn et al., 2018; Graham et al., 2005; Lee et al., 2024) and, therefore, also can influence cash flow.

As no studies have examined accrual accounting EM before and after the introduction of recapitalization suspension in Italy in response to the negative pandemic effects, this study analyses EM studies during a financial crisis and especially the COVID-19 pandemic. Indeed, the governmental policy of the suspending the recapitalization requirement was enacted in the context of a financial crisis, namely, during the pandemic, in response to its negative effects on the ability of firms to respect and understand the going concern assumption. Reviewing this literature, however will help to compare and contrast the results of this paper before and after the introduction of the regulation during the pandemic to the results from studies that focused on the same period but without considering the introduction of any new regulation.

Several scholars have questioned the effects of economic crises on EM. The topic has been extensively investigated, especially in family firms, and future research in this area was deemed to be essential (Naz et al., 2024).

The results achieved by academics do not always agree. A number of studies have demonstrated that crisis contexts provide incentives to managers to increase EM. Bornemann et al. (2012) found that, between 1997 and 2009, EM through 340f reserves became more relevant during financial crises by presenting a stable income stream, which is an important point to address during crises, particularly for commercial banks. Graham et al. (2005) conducted surveys and interviews with more than 400 executives to identify the factors that drive reported earnings and disclosure decisions and found that, among other results, that EM is more prevalent in bad times. Habib et al. (2013) examined managerial EM practices of financially distressed firms and considered a sample consisting of NZX-listed firms from 2000 to 2011. They found that managers of distressed firms engaged in more income-decreasing EM practices than did their healthy firm counterparts. Saleh and Ahmed (2005) considered similar issues for a sample of 153 firms listed on the KLSE during the Asian financial crisis.

Still, other studies have shown an inverse relationship between economic crises and EM. Cimini (2015) investigated the effects of the 2008 financial crisis on EM by sampling non-financial listed companies from 15 EU countries. His research found a decrease in EM since the beginning of that crisis. Filip and Raffournier (2014) achieved the same result and concluded that the reasons for this phenomenon could be found in “conservatism”, in the monitoring conducted by auditors (Ming Chia et al., 2007), in the higher quality of reporting required in times of crisis, in a greater tolerance for negative operating results during economic crises, and in the increased risk of litigation. Such findings argue again the belief that managers increase opportunistic behaviours during crises to achieve certain goals (Iatridis and Dimitras, 2013), such as having a “big bath” to clean up their balance sheets in anticipation of having an earnings boost after the crisis. Ma and Song (2016) also observed that EM becomes weaker during times of crisis.

Strobl (2013) and Campa (2019) offered possible explanations for such contradictory findings. Strobl found a relationship that depends on multiple aspects by considering an agency model of a multi-asset economy with risk-averse investors where managers can misstate companies’ earnings, Strobl showed that “the manager’s incentive to overstate a firm’s earnings is inversely related to the correlation between that firm’s earnings, and market: managers of firms whose earnings are more strongly correlated with the market during periods of economic expansion are more inclined to engage in manipulation during periods of recession, and vice versa”. Campa (2019) studied the different gradations of EM between listed and unlisted firms and explained the different EM practices based on the level of firm indebtedness. In high levels of indebtedness, there was a more extensive use of income-increasing EM behaviours among listed firms than among unlisted ones (Campa, 2019).

Table 1 below summarizes the contradictory findings of scholars who have investigated the effects of financial crises on EM.

Turning now to the COVID-19 pandemic, several specific studies have examined EM in different areas of the world, with the majority providing evidence of an EM increase. Šušak (2020) examined Croatian-listed companies and the effects of regulatory changes (introduced after the pandemic) on EM policies. That evidence indicated a positive relationship between financial statements’ lack of timeliness and EM activities. In the Eastern world, Yan et al. (2022) analysed how the pandemic affected the EM of Chinese companies. These researchers reported how the companies that were surveyed developed an increase in both accrual-based and real EM policies. Rahman et al. (2023) achieved the same results by showing that both family-owned and non-family-owned listed companies were equally affected. Ryu and Chae (2022) observed the same phenomenon for the distribution and services sector in Korea. Yassin et al. (2022) observed how, during the pandemic, the revenue standard was used as a tool for EM at both IFRS and US GAAP levels. Khanchel and Lassoued (2022) investigated whether a difference in intensity emerged in EM between socially responsible companies (SRCs) and non-SRCs during the COVID-19 pandemic. They found that the observed phenomenon was more pronounced in SRCs during the pandemic. Garfatta et al. (2023) studied the implications for Tunisian-listed companies and showed that companies adopted both accrual EM and real EM on a large scale during the reporting period; thus, the pandemic provided fertile ground for this type of strategy. More recently, Taylor et al. (2023) studied a sample of European banks to validate the increase in EM found by the majority of scholars. In addition, da Silva Flores et al. (2023) compared and contrasted EM in U.S. and Brazilian listed companies during the pandemic and found that companies in both Brazil and the U.S. practised EM more during the pandemic than during other crises. The researchers identified this action as a trend reversal from previous crises and found its justification in the analysed companies’ attempt to postpone the negative effects of the crisis on their balance sheets to the greatest extent possible.

While also studying the U.S., Liu and Sun (2022) focused on the impact of the COVID-19 pandemic on EM, but observed a decrease in discretionary accruals following an increase in absolute accruals between 2019 and 2020. The authors concluded that companies lowered their accruals to inflate their income in subsequent years. Uddin (2023) observed that non-financial companies listed in Bangladesh showed a lower incidence of EM policies. In terms of global observations, Ali et al. (2022) analysed a sample of companies from the G-12 countries and found that the incidence of EM decreased during the pandemic.

Despite some contradictory results, most of the previous studies have found that during the COVID-19 pandemic, companies increased EM practices to better weather the crisis period. Table 2 below summarizes the findings for the effect on EM produced by the pandemic crisis.

As previous research has indicated, the effects of financial crises on EM have shown contradictory results. Some studies have provided evidence that EM decreased during and after crises, while other studies have shown evidence of an EM increase. For the recent pandemic, the literature seems to suggest a general increase in EM. When examining the motivations behind these findings, no studies have theorised on the impact on EM of government reforms that were adopted in the context of the COVID-19 pandemic. Most of the motivations were based on the intention to mitigate the negative effects of the pandemic on their operating performance by reporting better performance (Ali et al., 2022) or, thanks to the so-called “big bath,” to postpone the positive performance to the end of the crisis (Garfatta et al., 2023). To fill this gap in the literature, this paper studies the effects of a government measure that suspended the recapitalization obligation to maintain the going concern assumption for EM.

Knowing this, we believe that the introduction of a policy to help firms maintain the going concern assumption in response to the pandemic might have reversed this EM trend, thereby leading to a reduction in management’s decision to manipulate earnings. This trend is particularly evident if the policy reduced the agency costs and the severity of the legal requirements and, in particular, if the same policy regarded the minimum capitalization obligations to respect in case of losses. As the book value of equity is directly correlated with and dependent on the reported earnings, a governmental measure that can alleviate the severity of possible capital erosion in a crisis could have a positive impact on management’s decision to manipulate earnings. After the introduction of such a governmental policy, it is far simpler to respect the going concern assumption and the minimum capital provisions required by law; therefore, management’s decision to manipulate earnings should be less likely.

Isomorphism (DiMaggio and Powell, 1983), which is a fundamental concept of new institutional theory (Rudko et al., 2024), has enabled this paper to extend the results from emerging countries, where corporate governance mechanisms are less effective in obstructing EM, to the developed countries (Almarayeh et al., 2022; Bao and Lewellyn, 2017). The ability of a new regulation to reduce agency costs (Jensen, 2005) is still another factor that indicates that less severe governmental policies do matter, as they can reduce agency costs and, in turn, also EM behaviour.

Based on these arguments, our hypothesis for this paper is as follows:

To test the hypothesis, we focused on the Italian context, where the governmental measure of the suspension of recapitalization obligation was introduced. For the period analysed, EM was detected in 2018 and 2019 (the pre-suspension period) and also 2020 and 2021 (the post-suspension period). The year 2021 was selected as the cut-off year, as it is the last year with data available for all the selected companies. The historical series used 2018 as its starting point to maintain the appropriate balance between the selected observed periods.

EM was detected using non-listed companies as a reference; in our experience, these companies are more inclined to engage in manipulation during periods of recession when the choice to recapitalize is far more recurrent. For these entities, the internal audit function is less structured with respect to the listed entities and thus, given a weaker internal audit function, the EM practices are more common (Alzoubi, 2019; Ghaleb et al., 2020; Prawitt et al., 2009), as high-quality auditing has an inhibitory effect on accrual-based EM, but has no inhibitory effect on real EM (Yan et al., 2022).

To detect EM, we used Larcker and Richardson’s (2004) model for the main analysis.

In our sensitivity analyses, several other approaches that are widely used in the literature (Leuz and Wysocki, 2016) were adopted to test the robustness of our findings.

The main model can thus be specified in the following equation:

where:

• TA_it is the total accruals over the estimation period;

• A_it−1 is the lagged total asset of the firm i;

• ∆REV_it is the change in firm i revenues during the event period;

• ∆REC_it is the change in firm i receivables during the event period;

• PPE_it is the gross property, plant, and equipment of firm i during the event period;

• BM_it is the boot-to-market ratio represented by the book value of equity for the unlisted firms; and

• CFO_it is the cash flow from operations.

The residuals of these models were considered as our measure of abnormal accruals. Their absolute value is thus a measure of EM (Hribar and Craig Nichols, 2007; Larcker and Richardson, 2004). They were estimated for the period before (i.e. 2018–2019) and after (i.e. 2020–2021) the introduction of the suspension of recapitalization obligation. We expected to find lower average absolute values of abnormal accruals after the introduction of the suspension of recapitalization obligation than before that suspension. In such a case, the model would provide evidence of a decrease in management’s decisions to manipulate earnings after the introduction of this transitory reform.

According to Callao and Jarne (2010, pp. 167–168), “this model starts from the modified version of the Jones (1991) model proposed by Dechow et al. (1995) and attempts to improve it by including BM and CFO. These variables are included because it is likely that incentives to manage earnings vary in response to growth opportunities (BM is included as a proxy for expected growth in the firm’s operations) and current operating performance (measured by CFO)”.

All the data necessary to implement our models were downloaded from the AIDA database, the Italian Digital Database of Companies and provided by Bureau van Dijk and commonly used in accounting and management studies (Calabrò et al., 2018). No data were hand-collected from the annual reports of the entities being analysed thus avoiding any bias in the data collection.

To build our sample using the AIDA database, we started with the legally active companies for a set of 1.5 million Italian companies. We then entered several filters. First, all companies that did not correspond to a legal form of a corporation were excluded (112,423 firms). For the core business nature, only non-financial companies were considered for our sample, a choice that excluded 62,583 financial entities. Such an exclusion is typical of EM research and was undertaken to make the sample more homogeneous and not biased by composition of the sample composition (Leuz et al., 2003). Minor companies (revenues under 8.8 million, assets under 4.4 million, and employees under 50 units) were not considered due to the lack of separation between ownership and control that is typical according to the agency theory that this paper sought extend. As a result, 1.3 million companies were excluded. Selecting only unlisted companies led to a sample of 14,410 companies. After excluding units for which there was no available data (6 companies), we arrived at a final sample of 14,404 companies. When considering fiscal years 2018–2019 for the period before and 2020–2021 for the period after the suspension of recapitalization obligation, our dataset included 57,616 firm-year observations. Table 3 below summarizes our sample selection strategy and provides further insights into the geographical distribution of the entities that were analysed.

Once the sample was selected, we downloaded financial and other data for each company included in the sample for 2017–2021 (the period under study was 2018–2021) from the AIDA database. Table 4 below provides the main descriptive statistics of the variables that were used on the models to detect EM.

As Table 4 clearly demonstrates, the sample includes loss-making companies, as indicated by a negative minimum for both equity and net income. For the dimensions of the companies in the sample, the mean value of equity was 47852.60 and the total assets’ mean value was 115503.30. Interestingly, the cash flow minimum mean was far below 0 (−1,012,868) while its mean value was only 876.96. This result could be due to the historical period under review that was marked by a deep crisis. Again, for this reason, receivables write-down (whose mean value was 319.25) and other depreciations (whose mean value was 1176.02) play an important role in the accruals’ assessment, accounting for 47.75% of the net income (the mean value being 3131.42).

Table 5 summarizes our findings by using Larcker and Richardson’s (2004) model.

Assuming the absolute value of the residuals of our model, estimated both before and after the suspension of recapitalization obligation, our proxy of abnormal accruals (i.e. EM), the results validated our research hypothesis. The absolute value reduction of the accruals can thus be envisaged from the absolute value of not only the mean (0.07 for 2018–2019 and 0.00 for 2020–2021), but also the median (0.06 for 2018–2019 and 0.01 for 2020–2021), offering evidence of a reduction in EM.

To test the robustness of these findings from the main analysis, we conducted several robustness checks. First, we tested the research hypothesis by using total accruals as a proxy of EM (Healy, 1985). Secondly, we adopted Jones’s (1991) alternative modified model, as adopted by Dechow et al. (1995). Third, we used the ratio analysis commonly adopted to detect EM as an alternative to regression models (Leuz et al., 2003). In particular, we were able to calculate the following indexes:

where:

• σ(OE_it) is the standard deviation of the operating earnings of entity i at the end of fiscal year t; and

• σ(CF_it) is the standard deviation of cash flow from operations of entity i at the end of fiscal year t.

where:

• abs(TA_it) is the absolute value of total accruals of entity i at the end of fiscal year t; and

• abs(CF_it) is the absolute value of cash flow from operations of entity i at the end of fiscal year t.

Our hypothesis was thus validated if the first index increases and the second index decreases. In the first case (R₁), an index increase suggests less smoothed earnings and thus, less EM. In the second case (R₂), a decrease in the index suggests earnings with lower discretionary components and, thus, a decrease in EM. The findings of these tests are tabulated in Panels a), b), and c) of Table 6.

These results suggest that our hypothesis is valid. In particular, Panel a) highlights a reduction in total accruals. Actually, their mean value decreased over the two periods analysed (from 2885.70 to 1623.22). As total accruals are considered a proxy of EM in this particular model (Healy, 1985), the results validated our hypothesis. In addition, the standard deviation value increased over the two periods (from 38790.71 to 47819.31), highlighting a reduction of earnings smoothing. Furthermore, Panel b) underscored a reduction in EM. Jones (1991) assumed abnormal accruals to be the discretionary components of total accruals. In particular, the absolute mean value of abnormal accruals decreased over the two periods (from 0.08 to 0.01), suggesting a decrease in EM. In terms of the indices, Panel c) suggests that the mean value of the first indicator (R₁) increased over the two periods, from 58.71 to 62.84, suggesting a decrease of EM.

This result is highly interesting for our research. Actually, earnings smoothing can be used to avoid recapitalization (in a system with a recapitalization obligation). Without this kind of obligation, companies are less encouraged to ensure minimum capital through smoothed earnings. In the second case (R₂), the value of the index decreased from 125.91 in the first period to 121.16 in the second period, suggesting a reduction of EM.

As the main analysis and sensitivity test results clearly indicate, the empirical evidence suggests that the EM decreased in 2020–2021 compared to 2018–2019. Each of the main methods for identifying EM also yielded coherent results in our research, despite the fact that the literature has shown that EM increased during the pandemic crisis (da Silva Flores et al., 2023; Garfatta et al., 2023; Rahman et al., 2023; Ryu and Chae, 2022; Taylor et al., 2023; Yan et al., 2022). This reduction might be due to the less severe restrictions placed on recapitalization requirements that overcame the main motivations provided by the literature to support the increase of EM, that is, the management intention to mitigate the negative effects of the pandemic by reporting a better performance (Ali et al., 2022) or to postpone the positive performance to the end of the crisis, thanks to the so-called “big bath” (Garfatta et al., 2023).

Although there is no homogeneity for the results of studies on EM during financial crises in general, studies that have specifically concerned the COVID-19 crisis seem to show that EM practices increased during the pandemic. This article thus analysed how the introduction of the suspension of the recapitalization obligation reversed this situation, leading to a decrease in EM practices. An analysis of a sample of 14,404 unlisted non-financial Italian companies in 2018–2021 validated the hypothesis that during the pandemic, EM decreased after the introduction of a governmental policy to reduce the severity of legal requirements and facilitated the going concern assumption. The conviction that the EM decrease mainly stemmed from the fact that the obligation to have a certain level of capital was a major reason for manipulating earnings, disappeared with the suspension of the recapitalization obligation.

These results are absolutely coherent with the assumption of institutional theory, according to which regulative elements and activities (i.e. EM in our case) are useful for delivering stability to each organization. From an institutional perspective, variations in institutional characteristics, such as values, norms, and solid governance monitoring, can be a source of pressure that can affect managers’ incentives (Ratu and Rahajeng, 2024, p. 350), and in particular EM, to guarantee a firm’s stability. The introduction of less severe governmental policies on recapitalization obligation reduces the governance monitoring, so managers may have an incentive to avoid underestimating loan loss provisions and/or depreciations and amortizations of expenses, thereby reducing EM with respect to the period before their introduction. These results also provide clear insights that EM behaviour is affected by the issuance of a new regulation that can obstruct EM, to the extent that it is able to reduce agency costs (Jensen, 2005) and empower activities and actors (Scott, 2013, p. 58). This conclusion extends the results of the latest research that investigated this ability, particularly in emerging countries (Almarayeh et al., 2022; Bao and Lewellyn, 2017). These results further show how EM is influenced by not only the quality of accounting standards, but also by government activities and decisions. This paper thereby contributes to actual business practice by identifying how management practices, like accounting standards, affect the quality of financial reporting.

Still this paper is not free of limitations. The most intrusive limitation is that there have been many government measures, so one cannot exclude a priori the potential “combined effects” of them. Also at the regional level, there were different measures adopted by different countries. Another limitation regards the focus of this paper on a single dimension of accounting quality, that is, EM. Future studies might control for these “combined effects” or expand the topic further by studying the impact of government measures on other dimension of accounting quality, such as conservatism, by considering the effects that those policies might have at regional levels or more generally on other countries or listed companies.