Do IFRS convergence affects firm performance? Picturing Indian-listed manufacturing firms

Abstract

Purpose

The study has endeavored to assay the nexus between the converged version of the International Financial Reporting Standards (IFRS) on the performance of the Indian-listed manufacturing firms.

Design/methodology/approach

The study has randomly accessed the data of the Bombay Stock Exchange (BSE) listed Indian manufacturing firms using the Prowess IQ database. It has covered 2014–2016 as pre-IFRS and 2017–2020 as the post-IFRS convergence period. Moreover, the study has followed a longitudinal research design with cross-sectional time-series data and has used the difference-in-difference (DiD) technique to assess the effect of the IFRS convergence on firm performance (FP).

Findings

The results have indicated that the adoption of the Indian Accounting Standards (Ind AS) has unlikely reported better FP. It has concurred policy implications as full adoption rather than convergence could reap the benefits of the IFRS.

Originality/value

It has contributed to the existing body of knowledge by assaying the effect of the IFRS convergence on FP in developing economies like India using the DiD methodology. The study is an original piece of research and is free from plagiarism.

1. Introduction

The International Accounting Standards Board (IASB) has formulated the International Financial Reporting Standards (IFRS) to harmonize diverse accounting practices worldwide to ensure adequate accounting information to the users of the financial statements (FSs). The significant drivers of the worldwide adoption of the IFRS have been attributed to the globalized capital markets and the rapid growth of international trade. Currently, 166 jurisdictions, including India, have adopted the IFRS for preparing FSs (IFRS Foundation, 2022). Interestingly, India has followed the convergence route through carve-in/out in the existing IFRS and has adopted the modified version of the IFRS, i.e. the Indian Accounting Standards (Ind AS) for financial reporting. Accounting standards have been intended to implement by the entities operating in a particular territory. The IFRS, as a global standard, likely has overlooked environmental factors such as the culture and customs of the particular territory. Thus, the standard-setters have formulated the accounting standards considering these factors and the competencies of preparers and auditors of FSs (Bhattacharyya, 2013). Again, the convergence of the local Generally Accepted Accounting Principles (GAAP) in line with the IFRS would likely to constrain the efficacy of the IFRS (Ball, 2016). However, the underlying philosophy of introducing the IFRS has to make FSs understandable, comparable, relevant and reliable amongst the financial markets to improve the quality of accounting disclosure to reduce information asymmetry for economic decision-making (Benkraiem et al., 2022). Consequently, a significant reduction in the information asymmetry has reduced the cost of capital, improved investment efficiency, enhanced analyst following, increased cross-border acquisitions, reduced earnings management and contributed to the country's economic growth (Wijayana and Gray, 2019; Akisik et al., 2020). Improvement in the disclosure and quality of accounting information has enhanced investors' assessment of the firm performance (FP) (as measured by an increase or decrease in the profit earning ability over a given period) and willingness to invest more as a more favorable result of the assessment has attracted more investments (Ofoegbu and Odoemelam, 2018).

Prior studies have documented inconclusive evidence with the IFRS adoption and FP (Ali et al., 2016; Miah, 2020). It has been argued that the IFRS adoption has enhanced the firm's profitability in developed and developing economies (O'Connell and Sullivan, 2008; Cordazzo, 2013). On the contrary, research has also validated the insignificant or no impact of the IFRS adoption on the FP (Păşcan and Ţurcaş, 2012). The literature has suggested that studies addressing the effect of the IFRS adoption on FP in developing economies are likely in deficit with a few exceptions (e.g., Agyei-Boahpeah et al., 2020). Moreover, return on assets (ROA) has been recognized as a better proxy to gauge the FP vis-à-vis other accounting-based and market-based performance measures and has been widely used in the literature (Miah, 2020). Interestingly, the literature has shown the poor performance of the Indian manufacturing sector due to inadequate foreign direct investment and infrastructural support and stagnation in the employment generation are the prominent factors (Sarkar et al., 2021). However, India could provide a distinctive background to explore the implications of the IFRS convergence on FP. Notably, unlike Australia, Canada and European Union (EU) countries, India, a developing economy, has converged her native reporting standards with the IFRS, indicating significant variations in reporting from the IFRS (Krishnan, 2018; Tawiah, 2020; Bansal and Garg, 2021). Interestingly, as opposed to Western economies, India has been dominated significantly by the government and family ownership firms having businesses yielding significant differences in compliance with the IFRS (Abdelqader et al., 2021). Moreover, coercive isomorphism exerted by informal market forces rather than formal supervisory mechanisms may hinder the IFRS convergence process (Silva et al., 2021). An extensive review of the Ind AS-related studies has shown research has been carried out, and a few of those include expectation gap analysis of practitioners on the IFRS convergence (Deb and Das, 2018), corporate reporting practices (Deb et al., 2021), accounting quality (Bansal and Garg, 2021), earnings management (Bansal, 2022) and the impact study of the Ind AS carried out by the Institute of Chartered Accountants of India (ICAI) (ICAI, 2018). The survey of related studies has indicated that the literature on the identified research problem is likely in scant, and current research would attempt to replenish the gap based on empirical evidence.

Contribution of the study in the literature is multiple. At first, it has significantly aided to the debate on the relative costs and benefits of the IFRS implementation in developing countries. Second, it has highlighted FP has decreased due to the adoption of the converged IFRS, contradicting the literature (Miah, 2020). In corollary, it has also likely to be concluded that the IFRS convergence would decrease accounting quality in general and disclosure in particular, thereby adversely affecting the FP. Finally, it has highlighted the impact of the IFRS convergence on FP, which could guide the developing economies in planning to converge their domestic standards in line with the IFRS.

The study has endeavored to assay the effect of the IFRS convergence on FP in the Indian-listed manufacturing firms.

The subsequent sections of the study have been framed as in Section 2, the theoretical framework and hypothesis development; in Section 3, the adopted methodology, Section 4 has incorporated results and discussion and Section 5, the study has derived its conclusion.

2. Theoretical framework and hypothesis development

2.1 Theoretical framework

In line with the emergence of the IFRS-related studies, the literature has also identified key theories to explicate the IFRS adoption in developing economies. The two notable macro-economic theories, namely the economic theory of networks (Katz and Shapiro, 1985) and isomorphism (DiMaggio and Powell, 1991), have explained the motives for adopting the IFRS in developing economies. According to the economic theory of networks, nations prefer to adopt the IFRS if two economies within the geographical region are the IFRS adopters. Again, if one nation has a close economic connection with an IFRS adopting country, the former is more likely to embrace the latter to facilitate multinational operations, thereby reducing the domestic biases faced by foreign investors (Ramanna and Sletten, 2009). Furthermore, as theory has postulated, countries would adopt the IFRS if the network benefits from the evolvement of the IFRS despite direct benefits from such standards being inferior to the domestic standards. On the other hand, three types of isomorphism have explained the IFRS adoption in the developing countries. Coercive isomorphism implies that nations would adopt the international standards and integrate their local accounting standards with the IFRS under formal or informal institutional forces (Zeghal and Mhedhbi, 2006). Second, the mimetic isomorphism has been referred to as the imitation of the IFRS adopting nations considered appropriate and successful (Hassan, 2008). Finally, in normative isomorphism, the levels of educational attainment in a country likely have compelled firms to shift towards adoption of the IFRS. The literature has identified four key theories, namely agency theory, signaling, political cost theory and capital need theory to define the magnitude of compliance with the global standards in developing countries. The choice of accounting practices and disclosure of financial information of the manager has been explained through agency theory, which has posited that the accounting choices and disclosure have been used to minimize the agency cost, thereby reducing information asymmetry between managers and shareholders. The IFRS compliance would likely indicate the restriction on the accounting choices and improvement in the extent of disclosure thus attributing to the existing agency cost to explain firms' behaviors towards the IFRS compliance. The proponents of the signaling theory have postulated that the market perceives firms as valuable, which has substantially minimized the information asymmetry for economic decision-making (Spence, 1973). Consequently, firms in the market have differentiated as high-quality by signaling investors adopting the best accounting policy that restricts accounting choices and enhance disclosures (Morris, 1987). So, the IFRS adoption would likely to signal investors about the improved disclosure to ease economic decision-making. Again, the capital need theory explains a firm's behavior to raise capital as cheaply as practicable. The literature has documented that firms tend to increase the quality of disclosure to inform the investors about their position and to increase the certainty of future cash flow (Craven and Marston, 1999). The IFRS being principle-based accounting standards would likely enhance compliance to acquire a cost-effective corporate capital and would likely contribute to improving the FP. Furthermore, value maximization theory has posited that a firm tends to maximize profit in the short-run and maximization of shareholders wealth in the long run (Jensen, 2001). The long-run wealth maximization would indicate the shareholder's wealth and the maximization of other financial claimants such as debt and warrant holders, which could be achieved by increasing long-run profitability. Consequently, the IFRS as high-quality accounting standards having adequate disclosure would provide significant accounting information to the stakeholders for maximizing long-term profitability and FP.

2.2 Literature review and hypothesis development

The literature has documented mixed evidence pertinent to the IFRS adoption and FP. The IFRS-adopted firms have reported higher profits vis-a-vis firms following domestic accounting standards (Ali et al., 2016), albeit firms that have reported detailed accounting disclosure have also shown enhanced profitability (Iatridis, 2008). Again, the mandatory IFRS adoption has channeled a substantial increase in the firms' net income amongst the EU countries (O'Connell and Sullivan, 2008). Furthermore, in the Chinese context, it has been reported that the IFRS implementation has improved the FP to a certain extent (Miah, 2020). The effect of the IFRS implementation on firm values in the African countries has reported that the full adoption of the IFRS rather than partial/modified adoption has improved the FP (Agyei-Boahpeah et al., 2020). Interestingly, FP has been more pronounced in countries with a greater commitment to the rule of law. Again, differences in the reported FP under the Italian GAAP and the IFRS, show that firms reporting under the IFRS have documented improved FP (Cordazzo, 2013) while converged IFRS-adopted firms have reported insignificant impact on the FP (Păşcan and Ţurcaş, 2012).The IFRS adoption in the New Zealand has reported a significant positive impact on common financial ratios such as ROA, ROE, leverage and return on sales of the firms (Stent et al., 2010). Again, the literature has argued the manifold benefits of the IFRS adoption due to improved disclosure (Benkraiem et al., 2022). The literature has reported that higher transparency has decreased the cost of capital, enhanced capital efficiency, promoted cross-border acquisitions, reduced income smoothing, and led to the country's economic growth (Wijayana and Gray, 2019; Akisik et al., 2020).

In the Indian context, firms reporting under the converged IFRS have generally exhibited poorer accounting quality and decreased FP, as measured by earnings per share (EPS) (Bansal and Garg, 2021). It has also been argued that the Indian capital market has responded adversely due to the Ind AS implementation, as firms have reported increased costs of both equity and debt, increased information asymmetry and reported a reduction in market liquidity (Bansal, 2022). Interestingly, the literature has also highlighted that the propensity of earnings management has enhanced among the IFRS-implementing Indian firms (Himanshu and Singh, 2021). Based on the literature, the present study has expected a negative relationship of the IFRS convergence with the FP, and accordingly, it has framed the hypothesis as follows:H1.

IFRS convergence has a detrimental influence on FP.

3. Research methodology

This section has been developed incorporating the subsequent sub-sections.

3.1 Research design

The study has followed a longitudinal research design with cross-sectional time-series data.

3.2 Method

The Ind AS implementation has not happened in isolation, as the government has announced demonetization and introduced the Goods and Services Tax (GST) in 2016 and 2017, respectively. Those economic shocks have made the Ind AS implementation challenging to observe the pure change in the performance among the Ind AS adopting firms. The present study has employed the difference-in-difference (DiD) technique to exude the effect of concurrent economic shocks ensuring the change occurred only due to the Ind AS implementation. The DiD technique has been considered quasi-experimental and an advanced econometrics tool which could help isolate the effect of the particular economic shock on the concerned variable. The DiD process has required data from two groups of two different periods. It has categorized the first group of firms as treatment firms' (firms mandatorily adopted the Ind AS w. e. f. The April, 2016) and the second group as control firms' (firms' exempted from the Ind AS adoption). The study has divided the period between pre-Ind AS (2014–2016), i.e., the period before the mandatory Ind AS adoption and post-Ind AS adoption (2017–2020), which has implied the period after the mandatory Ind AS implementation for the analysis.

The Ind AS has been mandatorily implemented in the two phases considering the net worth of more than INR 5 bn and INR 2.5 bn., respectively. It has been observed in the annual reports that initially, the firms in both phases did not implement the Ind AS mandatorily, but in the later periods, i.e., in FY 2018, FY 2019 and FY 2020, the sample firms have mandatorily implemented the Ind AS. Consequently, the firms in both the phases could not be considered the control group as it has received treatments although in the later periods. As a result, the study could not trace any non-Ind AS adopting firms to be considered the control group from Phase 1 and Phase 2 during the study period. It has considered the Ind AS exempted firms a control group with a net worth less than INR 2.5 bn and firms in the first phase of the Ind AS implementation as a treatment group with a net worth of more than INR 5 bn. Moreover, the DiD technique has allowed for the comparison of two groups of firms for the two different periods although the current study has identified three groups of sample firms, namely, firms adopting the Ind AS in the first phase, in the second phase and the Ind AS exempted firms for the comparison. Accordingly, it has excluded the Phase 2 implementation year from the sample and has compared the performance between Phase 1 and Ind AS-exempted firms as treatment and control firms, respectively. Furthermore, it has also excluded the sample firms voluntarily adopting the Ind AS since it could likely impact the findings' validity.

3.3 Data and sample selection

The study has collected data using the Centre for Monitoring Indian Economy (CMIE) Prowess IQ database covering 2014–2016 as pre-IFRS convergence (before the Ind AS adoption period) and the 2017–2020 post-IFRS convergence (after the Ind AS adoption period). It has randomly analyzed the BSE-listed Indian manufacturing firms. The Ind AS has been mandated in two phases considering the firms' net worth. In the Phase 1 (April, 2016 onwards), firms having a net worth of more than INR 5 bn have mandatorily adopted the Ind AS. During the Phase 2 of the Ind AS implementation (April, 2017 onwards), it has also been obligatory for firms having a net worth of more than INR 2.5 bn. However, firms with a net worth of less than INR 2.5 bn have been exempted from mandatory adoption of the Ind AS. The sample selection procedure for treatment and control firms has been presented in Table 1.

3.4 Study variables

The present study has identified the variables based on the related literature and has categorized as the outcome, predictor and control variables. The description of the study variables has been presented in Table 2.

3.5 Model estimation

The present study has developed the model following Fauver et al. (2017) and Agyei-Boahpeah et al. (2020) for assaying the effect of the IFRS convergence on FP. The model has appeared as follows:$Y_{it}=a+{POST}_{{IndAS}^t}+{Control}_{it}+{\epsilon }_{it}$where

• Y_it = Dependent variable.

• POST_IndASt = post-IFRS convergence timeline.

• Control_it = Control variables.

• ɛ_it = Error term.

The extended equation for gauging the effect of IFRS convergence on FP is mentioned as follows:${ROA}_{it}=\alpha +{\beta }_1{POST}_t+{\beta }_2{IFRS*POST}_{it}+{\beta }_3{INV}_{it}+{\beta }_4{REC}_{it}+{\beta }_5{LEV}_{it}+{\beta }_6{CF}_{it}+{\beta }_7{SIZE}_{it}+{\epsilon }_{it}$where ROA has represented FP and POST, a dummy variable would indicate 1 for the post-IFRS convergence timeline and 0 for the pre-IFRS convergence period. IFRS dummy variable has indicated “1” for treatment firms and “0” for control firms. The interaction effect of the variable IFRS*POST would indicate the influence of IFRS convergence on treatment firms' performance.

3.6 Inferential statistics

After running the ordinary least square regression, analysis has been initiated with the identification of variance inflation factor (VIF). The test has highlighted VIF less than 2 for each study variable. The mean VIF has shown a value of 1.05 (see Appendix), validating that variables have been free from multicollinearity. It has also used the Wooldridge test and Breusch–Pagan/Cook–Weisberg test to identify the presence of autocorrelation and heteroskedasticity in the panel data. The analysis has indicated that the dataset contains autocorrelation and heteroskedasticity issue (see Appendix). The study has used the vector correction error (VCE) technique with cluster robust standards errors in the panel data regression to mitigate the effect of possible biases of the regression outcomes. Moreover, it has tested the regression model using the Hausman test to identify between the fixed and random effect models (see, Appendix). The DiD technique and panel data regression have provided consistent results, which has been presented in Section 4.

4. Result and discussion

A comparative descriptive statistics of the mean estimates of the variables have been presented in Table 3. The results have indicated that the treatment firms have highlighted a higher mean value of ROA (0.051) than control firms (−0.020) during the pre-Ind AS adoption period. It has also compared the mean values of ROA between the treatment firms (0.049) and control firms (0.005), following the post-IFRS convergence period. Compared to the pre-IFRS convergence timeline, the mean ROA of the former has substantially declined from 0.051 to 0.049 during the post-IFRS convergence period. It has provided early evidence that the firms' profitability has reduced to a certain extent after the post-IFRS convergence period. The negative change in the mean values of the variables INV, REC, CF and LEV among treatment firms have indicated a decrease in the ROA during the post-IFRS convergence period.

Additional analysis has been performed using the DiD technique with linear regression to assess the differential impact between treatment and control firms. It has analyzed the impact before and after the Ind AS adoption period. In Table 4, the result of the DiD estimation has highlighted significant difference between the pre-IFRS convergence period (0.046***) and the post-IFRS convergence period (0.024***) for the treatment and control firms, respectively. Treatment firms have reported a significant decrease in performance, as highlighted by the coefficient change from 0.016 to 0.014 following the IFRS convergence period. Moreover, the outcome of the DiD estimation (−0.022***) has revealed that the performance of the treatment firms has significantly reduced during pre-IFRS convergence period. The test results have validated that firms shifting to the Ind AS from the local GAAP have reported declining performance.

The outcome of the DiD estimation with the fixed-effect model has been presented in Table 5. The coefficient (−0.023**) of the DiD estimation has indicated significant difference in the performance of the firms during post-IFRS convergence timeline. The results have been derived after controlling the firm and the time effect. The test results have implied that treatment firms have reported a substantial reduction in performance compared to the control firms following the IFRS convergence. The post-estimation of the results of the DiD with bootstrap inference has been presented in Figure 1.

Figure 1 has shown graphical diagnostics for parallel trends of the observed means and linear trends, respectively. Observed means have highlighted that treatment and control firms have exhibited a declining trend in the mean values of ROA during the post-Ind AS adoption period. Moreover, observed means have indicated that control firms have reported higher variation in the ROA during the post-Ind AS adoption period. However, the linear trend has validated the differences as treatment firms have a declining trend than control firms following the Ind AS adoption. The results have shown that firms shifting from domestic accounting standards to the Ind AS have reported a significant decline in performance.

In Table 6, the results of the panel data regression have been presented. It has performed the Hausman test to select an appropriate model between fixed and random effects. The test has rejected the null hypothesis for choosing the fixed effect model (χ² 717.91; p < 0.05) (See Appendix). Thus, the analysis and interpretation of the results have been performed using the fixed effect regression model. The results have highlighted a significant positive, although a weaker POST (0.022*) relationship with ROA. It has implied that firms' profitability has improved to a certain extent following Ind AS adoption period.

Interestingly, the significant negative coefficient of the interaction term of IFRS*POST (−0.023**) has indicated that treatment firms have reported a significant decline in the FP after the Ind AS adoption period. As documented in the literature, the findings could be attributed to the convergence mechanism and interpretation of the accounting standards (Bansal and Garg, 2021). Taking the country's economic and legal settings into cognizance, it has adopted the modified version of the IFRS, and hence the efficacy of the original version of the IFRS would likely have diluted (Agyei-Boahpeah et al., 2020). Indian weak corporate governance and lax legal enforcement (Narayanaswami et al., 2011) could have probably hindered the benefits of the IFRS convergence. The findings have been consistent with random effect regression without changing any inferences. Thus, the results have produced evidence to likely accept H1 and reported that the IFRS convergence in India has substantially reduced the FP.

The control variables INV, REC and CF have shown a positive relationship with ROA supporting the literature (Stent et al., 2010) and have contested as well (Miah, 2020). The results have reported that LEV has a significant negative relationship with ROA, in tune with the literature (Bansal and Garg, 2021). However, SIZE has highlighted a statistically insignificant relationship with ROA, confirming the literature (Cordazzo, 2013). The positive relationship of INV with ROA has indicated that firms with better inventory positions positively impact the profit earning capacity of the firms. It has also shown that an increase in REC positively impacts ROA. It has highlighted an established positive relationship between an increase in sales and firms' profitability. A significant negative relationship of LEV with ROA has indicated that the propensity of debt financing decreases firms' profitability in tune with the literature (Bansal, 2022). The CF has highlighted a positive impact on ROA although it has contradicted with the findings of recent literature (see Bansal and Garg, 2021). It has implied that firms with better cash flow generating capacity would likely have better investment opportunities, thereby increasing profitability. The SIZE has reported an insignificant relationship with ROA. It has contrasted the view that larger-size firms have more resources to invest, which would lead to better FP (Miah, 2020).

5. Conclusion

The study has empirically investigated whether the IFRS convergences in India affect the FP. It has analyzed the data from the Indian manufacturing firms using the DiD technique and reported that the FP has deteriorated substantially after the IFRS convergence. The DiD technique has controlled for the concurrent economic shocks and ensured that the change occurred in the FP only due to the IFRS convergence. The findings have corroborated the economic theory of networks, which has posited that if the network benefits from the expansion of the IFRS, countries would be more likely to adopt the international standards, even if the direct benefits from such standards have been inferior to those from locally developed standards. However, it has opposed the value maximization theory to maximize the wealth of the shareholders in the long run, thereby increasing the overall profitability of the firm. The findings have contradicted the literature that has found a positive impact of the IFRS adoption on FP (Miah, 2020). Per contra, it has followed literature reporting the IFRS convergence in India has significantly decreased the FP (Bansal and Garg, 2021). Prior studies have identified India having a weaker corporate governance mechanism (Narayanaswamy et al., 2012), and countries that have more robust enforcement mechanisms would reap the benefits from the IFRS adoption (Karampinis and Hevas, 2011). Inasmuch as countries adopt the IFRS to improve accounting quality the results have unlikely indicated any such precedence probably due to India's lack of expertise in the IFRS training and learning programs, poor information technology and inadequate infrastructural support for successful IFRS implementation (Sharma et al., 2017).

The study has acknowledged a few limitations. First, it has selected the sample from the Indian manufacturing firms to generalize the findings. Second, it has incorporated accounting-based measures of FP in preference to the market-based measures. Finally, the study could not establish a nexus between corporate governance attributes and the IFRS implementation.

The findings have practical implications for multiple stakeholders. The findings could help policymakers for adopting the IFRS rather than convergence as partial/modified standards would likely dilute the full potential of high-quality standards. The results could be helpful for developing economies planning to converge their domestic accounting standards in line with the IFRS to understand the unintended economic consequences such as decreased FP. Countries could use this methodology to examine how the IFRS convergence has affected the FP by controlling external economic shocks. Managers of the Indian firms could likely to access the report to understand how the IFRS convergence has increased information asymmetry and eventually decreases the FP. The regulators and standard setters could take initiatives for improving the disclosure of the FSs to attract both domestic and international investors as well.

Future research endeavors could establish a nexus between the effects of corporate governance on the Ind AS compliance and determine the implications of the Ind AS on the Indian capital market for assaying the flow of capital across the sectors. Comparative studies between India and other emerging countries could be conducted to assess variations of adoption/convergence of the IFRS. Research could also examine the impact of the IFRS adoption on FP incorporating the excluded variables of the current study, e.g. rule of law index ROE, Tobin Q, firm growth, firm age, auditor type and other macroeconomic variables. Moreover, research could be endeavored considering more sample firms from across the sectors for generalizing the findings.

Figure 1

Graphical diagnostics for parallel trends*

[Figure omitted. See PDF]

Table 1

Sample selection procedure*

Particulars	Treatment firms	Control firms
Total sample firms	1,593	647
Firms with mandatory Ind AS adoption	1,593	0
Less: firms with voluntary Ind AS adoption	0	104
Less: firms with missing data	471	137
Final sample size	1,122	406

Source(s): *Authors' calculation

Table 2

Description of the study variables*

Variable	Type of variable	Definition	Author(s)
ROA	Outcome	Profit after tax scaled by total assets	Stent et al. (2010), Miah (2020)
POST	Predictor	Indicator variable 1 for post-Ind AS period; 0 otherwise	Agyei-Boapeah (2020)
IFRS	Predictor	Indicator variable 1 for firms with Ind AS adoption; 0 otherwise	Bansal and Garg (2021)
INV	Control	Inventory divided by total assets	Păşcan and Ţurcaş (2012)
REC	Control	Receivables divided by total assets	Miah (2020)
CF	Control	Cash flow from operations divided by total assets	Stent et al. (2010), Cordazzo (2013), Agyei-Boahpeah et al. (2020)
LEV	Control	Total debt scaled by total assets
SIZE	Control	Natural logarithm of total assets

Source(s): *Authors' compilation

Table 3

Comparative descriptive statistics with mean value of the variables*

Variables	Pre-IFRS convergence period		Post-IFRS convergence period
	Treatment firms	Control firms	Treatment firms	Control firms
ROA	0.051	−0.020	0.049	0.005
INV	0.179	0.206	0.172	0.197
REC	0.171	0.211	0.166	0.213
LEV	0.494	0.821	0.464	0.800
CF	0.083	0.043	0.081	0.055
SIZE	8.983	6.255	9.269	6.424

Source(s): *Authors' calculation

Table 4

Difference-in-differences with linear regression*

Variable	Pre-IFRS convergence period			Post-IFRS convergence period
ROA	Treatment firms	Control firms	Difference (T-C)	Treatment firms	Control firms	Difference (T-C)	DiD
	0.016	−0.030	0.046***	0.014	−0.010	0.024***	−0.022***

Note(s): Significance level at ***p < 0.01,**p < 0.05 and *p < 0.1, respectively

Source(s): *Authors' calculation

Table 5

Difference-in-differences with wild-cluster bootstrap inference*

ROA	Coefficient	t-statistics	p-value	[95% conf. Interval]
ATET
IFRS
(1 vs 0)	−0.023**	−2.430	0.012	−0.042–0.005

Note(s): Significance level at ***p < 0.01, **p < 0.05 and *p < 0.1, respectively

ATET estimate adjusted for covariates, panel effects and time effects. The results from treatment and control firms have been obtained to test for substantial changes between pre- and post-IFRS convergence periods. For longitudinal data, it has employed DiD estimation using wild-cluster bootstrap inference with 1,000 replications using the Rademacher error weight

Source(s): *Authors' calculation

Table 6

Fixed effect vs random effect regression with ROA*

Fixed effect		Random effect
ROA	Coefficient	ROA	Coefficient
POST	0.022* (1.890)	POST	0.020** (1.970)
IFRS*POST	−0.023** (−2.400)	IFRS*POST	−0.023** (−2.330)
INV	0.140* (1.770)	INV	0.084*** (2.970)
REC	0.154*** (3.250)	REC	0.093*** (5.580)
LEV	−0.147** (−2.290)	LEV	−0.071*** (−3.610)
CF	0.208*** (3.740)	CF	0.297*** (5.970)
SIZE	−0.010 (−0.410)	SIZE	0.002 (1.080)
Constant	0.135 (0.600)	Constant	−0.025* (−1.710)
Observations	10,696	Observations	10,696
Adj. R2	0.010	Adj. R2	0.012
p-value	0.000	p-value	0.000

Source(s): *Authors' calculation

Table A1

Multicollinearity test (Variance Inflation Factor)*

Variable	VIF	1/VIF
LEV	1.03	0.971399
SIZE	1.11	0.904943
INV	1.03	0.970057
REC	1.07	0.933364
CF	1.03	0.974788
Mean VIF	1.05

Source(s): *Authors' Calculation

Table A2

Breusch-Pagan / Cook-Weisberg Test for Heteroskedasticity*

Model with dependent variable	χ2	p-value
ROA	32,053.24	0.000***

Note(s): Significance level at ***p < 0.01, **p < 0.05, *p < 0.1, respectively

Source(s): *Authors' Calculation

Table A3

Wooldridge Test for Autocorrelation in Panel Data*

Model with dependent variable	F-statistics	p-value
ROA	5.579	0.018**

Note(s): Significance level at ***p < 0.01, **p < 0.05, *p < 0.1, respectively

Source(s): *Authors' Calculation

Table A4

Hausman Test for Fixed Vs. Random effect*

Model with dependent variable	χ2	p-value
ROA	717.91	0.000**

Note(s): Significance level at ***p < 0.01, **p < 0.05, *p < 0.1, respectively

Source(s): *Authors' Calculation