Re-examining the corporate governance – Firm performance nexus: Fresh evidence from a causal mediation analysis

Figures & data

Figure 1. The interlinkages among CGQ, FRQ and firm performance.

Source: Authors’ depiction of the inter-linkages among CG, FRQ and firm performance (2023).

Figure 2. The mediating role of FRQ in the CGQ and firm performance relationship.

Source: Authors’ conceptual framework of the mediating role of FRQ in the CGQ – firm performance nexus (2023).

Table 1. Sample selection

Country of sampled firms	Number of non-financial firms whose annual reports data were sourced from Africanfinancials and Machameratios databases for the study period	Number of firms with missing annual reports data over the study period	Number of firms whose missing annual reports data were deemed inadequate (i.e., having less than half of the annual reports data covering the study period), hence excluded from the sample	Number of firms with annual reports data deemed adequate (i.e., covering more than half of the study period) and therefore retained in the study sample
Ghana	5	0	0	5
Kenya	11	3	0	11
Malawi	2	2	0	2
Mauritius	5	3	0	5
Nigeria	40	8	3	37
Namibia	2	2	0	2
South Africa	35	13	0	35
Tanzania	3	0	0	3
Zambia	4	2	0	4
Total	107	33	3	104

Source: Authors’ compilation of annual reports from Africanfinancials and Machameratios websites.

Table 2. Measurement of variables used in the study’s models

Variable	Scale	Source	Expected Sign
Outcome Variable: ROA Return on Assets (Proxy for Firm-Performance/ Profitability)	This is measured by $\mathrm{R}\mathrm{O}{\mathrm{A}}_{i,t}$= $EBI{T}_{i,t}\hspace{0.17em}/T{A}_{i,t}$ Where: $EBI{T}_{i,t}$ = Profit before interest and tax for firm i in year t $T{A}_{i,t}$ = Total Assets for firm i in year t	Annual reports of firms
Mediator Variable: FRQ (proxied by Discretionary Accruals—i.e., DA)	This is measured using the Pae (2005) Discretionary Accrual’s Model;$N{A}_t$=${ }_{1}1/A_{t-1}$+${ }_2\mathrm{\Delta }\mathrm{R}\mathrm{e}{\mathrm{v}}_t/A_{t-1}$+${ }_3\mathrm{P}\mathrm{P}{\mathrm{E}}_t/A_{t-1}$ +${ }_4\mathrm{C}\mathrm{F}{\mathrm{O}}_t/A_{t-1}$ +${ }_5\mathrm{C}\mathrm{F}{\mathrm{O}}_{t-1}/A_{t-1}$ Where: $A_{t-1}$= total assets in the year t−1 $\mathrm{\Delta }\mathrm{R}\mathrm{e}{\mathrm{v}}_t$= the change in revenues from the preceding year $\mathrm{P}\mathrm{P}{\mathrm{E}}_t$= the gross value of property, plant and equipment in the year t $\mathrm{C}\mathrm{F}{\mathrm{O}}_t$= operating cashflows in the year t $\mathrm{C}\mathrm{F}{\mathrm{O}}_{t-1}$= operating cashflows in the year t−1 ${ }_1$, ${ }_2$, ${ }_3,{ }_4,{ }_5$= firm-specific parameters Estimates of the firm specific parameters are done via the model: $T{A}_t/A_{t-1}$ = ${\alpha }_{1}1/A_{t-1}$+${\alpha }_2\mathrm{\Delta }\mathrm{R}\mathrm{e}{\mathrm{v}}_t/A_{t-1}$+${\alpha }_3\mathrm{P}\mathrm{P}{\mathrm{E}}_t/A_{t-1}$ +${\alpha }_4\mathrm{C}\mathrm{F}{\mathrm{O}}_t/A_{t-1}$ +${\alpha }_5\mathrm{C}\mathrm{F}{\mathrm{O}}_{t-1}/A_{t-1}$ + ${\epsilon }_t$ Where: Total accruals (TA) is defined as income before extraordinary items and discontinued operations minus operating cash flows, that is, $T{A}_t$ = $NOP{I}_t$ - $CF{O}_t$ ${\epsilon }_t$ is discretionary accruals (DA) in the year t. And FRQ = (-|${\epsilon }_{\mathrm{t}}$ |)	Annual reports of firms	±
Exposure variable:
CGQ Corporate Governance Quality	This is measured as an index for firm i in year t, on the basis of firm-level disclosures on CG mechanisms. The CGQ index was constructed via means of rotated principal component analysis.	CACG cum OECD CG Guidelines, SEC CG Codes for respective countries, Annual reports of firms	±
Control Variables:
IFRS IFRS Adoption	This is a dummy variable measured as 1 since a firm’s adoption of IFRS and 0 otherwise.	Annual reports of firms	±
ASSTANG Asset Tangibility	This is measured as the ratio of tangible assets to total assets of firm i in year t.	Annual reports of firms	+
LEV Leverage	This is measured by the total liabilities to total assets. It is represented in logarithm form	Annual reports of firms	-
SIZE Firm size	This is measured as the logarithm of a firm’s total assets	Annual reports of firms	+
GRWTH Growth Opportunities	This is measured as the price-to-book ratio for firm i in year t.	Annual reports of firms	+
AGE Age of Firm	The is measured as the age of a firm from the date of listing on the stock market to the end of the sample period. It is represented in logarithm form.	Annual reports of firms	+
${\epsilon }_{i,t}$ Error term	The error term

Source: Authors’ Compilation, 2023.

Table 3. Descriptive statistics

	(1)	(2)	(3)	(4)	(5)
Variables	N	Mean	SD	Min	Max
SIZE	1,295	5.125	0.802	2.754	7.183
GRWTH	1,288	3.134	6.085	−36.47	96.98
IFRS	1,295	0.859	0.348	0	1
ASSTANG	1,254	0.400	0.237	0.00140	0.995
FRQ	1,295	0.0200	0.755	−2.268	26.10
CGQ	1,276	8.96e−09	1.000	−1.544	2.562
AGE	1,295	3.669	0.743	0	5.136
LEV	1,264	3.755	0.779	−1.926	4.604
ROA	1,295	6.031	15.52	−179.9	295.7
Number of groups	104	104	104	104	104

Notes: This table reports descriptive statistics based on aggregate samples of which the sizes may be various because of missing values. The variables are as defined in Table 2. For interpretation purposes, the descriptive statistics are calculated on the basis of levels with the exception of IFRS Adoption which was computed from a dummy scale, CGQ which was calculated as an index from normalized rotated principal component analysis, and Size, Age and Leverage were calculated on the basis of logarithmic form. The ROA, being the dependent variable in our model, was not transformed but allowed to retain its original form for 1) ease of interpretations, 2) because its histogram distribution appears normal.

Table 4. Correlation coefficients

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)		VIFS
(1) ROA	1.000
(2) CGQ	0.046*	1.000									1.18
	(0.099)
(3) FRQ	0.150***	0.050*	1.000								1.01
	(0.000)	(0.076)
(4) SIZE	0.161***	−0.150***	−0.007	1.000							1.17
	(0.000)	(0.000)	(0.810)
(5) GRWTH	0.226***	0.101***	−0.011	0.051*	1.000						1.10
	(0.000)	(0.000)	(0.700)	(0.065)
(6) AGE	0.042	0.105***	−0.009	0.133***	0.092***	1.000					1.18
	(0.130)	(0.000)	(0.734)	(0.000)	(0.001)
(7) LEV	−0.068**	0.057**	0.009	0.094***	0.040	0.083***	1.000				1.04
	(0.016)	(0.045)	(0.759)	(0.001)	(0.152)	(0.003)
(8) IFRS	0.001	−0.038	0.027	0.152***	−0.042	0.028	−0.046*	1.000			1.29
	(0.979)	(0.177)	(0.327)	(0.000)	(0.136)	(0.313)	(0.102)
(9) ASSTANG	0.070**	0.223***	−0.047*	−0.030	0.061**	0.130***	−0.026	−0.041	1.000		1.09
	(0.013)	(0.000)	(0.094)	(0.286)	(0.031)	(0.000)	(0.357)	(0.149)
(10) L.ROA	0.472***	0.047*	0.053*	0.172***	0.261***	0.061**	−0.084***	0.021	0.077***	1.000	1.14
	(0.000)	(0.110)	(0.068)	(0.000)	(0.000)	(0.037)	(0.004)	(0.462)	(0.009)

Notes: This table presents pairwise correlation coefficients which are based on samples of which the sizes may be various because of missing values. The variables are as defined Table 2. Asterisks indicate significance at 10% (*) 5% (**) and 1% (***).

Table 5. Baseline static and dynamic regression results of firm performance (i.e., ROA) determinants

	Static	Models	Dynamic	Models	Estimations
	(1)	(2)	(3)	(4)	(5)
Variables	OLS Estimator	FE Estimator	OLS Estimator	System GMM Estimator	FE Estimator
CGQ	0.882	0.948	0.580	0.428	1.209
	(0.593)	(1.380)	(0.384)	(0.469)	(1.136)
FRQ	2.746	2.336	2.399	1.283	2.333
	(2.833)	(2.802)	(2.714)	(3.651)	(2.811)
SIZE	1.999*	−4.978	1.155	1.963*	−6.977
	(1.048)	(7.121)	(0.839)	(1.100)	(8.185)
GRWTH	0.384*	0.178	0.242	0.117	0.177
	(0.201)	(0.140)	(0.192)	(0.195)	(0.164)
AGE	0.924	7.046	0.603	−0.0342	9.036*
	(0.713)	(4.802)	(0.486)	(0.622)	(4.948)
LEV	−2.193	−7.677***	−1.925	−2.852	−7.955***
	(1.745)	(1.910)	(1.369)	(2.272)	(1.963)
IFRS	2.710	2.659**	1.453	−2.153**	2.157
	(1.738)	(1.306)	(1.487)	(0.992)	(1.428)
ASSTANG	4.096	−0.810	2.890	2.658	−0.626
	(3.176)	(3.257)	(2.810)	(2.446)	(3.293)
L.ROA			0.358**	0.315***	0.0607
			(0.154)	(0.105)	(0.0988)
Constant	1,930***	2,137***	1,411***		2,240***
	(334.2)	(422.6)	(498.0)		(602.8)
Country fixed-effects	Yes	Yes	Yes	Yes	Yes
Year fixed-effects	Yes	Yes	Yes	Yes	Yes
Firm fixed-effects	Yes	No	Yes	Yes	No
Observations	1,210	1,210	1,115	1,115	1,115
R-squared	0.139	0.129	0.232		0.138
Number of groups	104	104	104	104	104
F-statistic	6.80***	7.34***	28.86***		10.20***
Wald chi-square statistic				392.71***

Robust standard errors in parentheses

*** p < 0.01, ** p < 0.05, * p < 0.1

Table 6. Static and dynamic regression results of firm performance (i.e., Tobin’s Q) determinants for robustness checks

	Static	Models	Dynamic	Model	Estimations
	(1)	(2)	(3)	(4)	(5)
Variables	OLS Estimator	FE Estimator	OLS Estimator	System GMM Estimator	FE Estimator
CGQ	0.0730	−0.000721	0.0835*	0.0409	0.0395
	(0.0483)	(0.0929)	(0.0460)	(0.0492)	(0.0949)
FRQ	0.0762	0.0622	0.0659	0.0253	0.0592
	(0.0886)	(0.0737)	(0.0799)	(0.0959)	(0.0679)
SIZE	0.146*	−0.667	0.166**	0.174	−0.690
	(0.0827)	(0.442)	(0.0828)	(0.109)	(0.530)
GRWTH	0.154***	0.0922**	0.143***	0.151**	0.0850**
	(0.0496)	(0.0383)	(0.0512)	(0.0744)	(0.0397)
AGE	0.0368	0.122	0.0450	0.0309	0.547*
	(0.104)	(0.296)	(0.0938)	(0.0661)	(0.279)
LEV	−0.0464	−0.000349	−0.0384	0.0140	−0.106
	(0.130)	(0.162)	(0.124)	(0.110)	(0.162)
IFRS	0.000207	0.0472	−0.0153	−0.222*	0.0683
	(0.166)	(0.158)	(0.150)	(0.128)	(0.170)
ASSTANG	0.506	0.191	0.557	0.661	0.220
	(0.353)	(0.229)	(0.350)	(0.404)	(0.236)
L.TobinsQ			0.0433	0.0205	0.0140
			(0.0427)	(0.0201)	(0.0155)
Constant	106.2***	103.9***	85.25***		113.7***
	(28.25)	(31.22)	(25.54)		(33.26)
Country fixed-effects	Yes	Yes	Yes	Yes	Yes
Year fixed-effects	Yes	Yes	Yes	Yes	Yes
Firm fixed-effects	Yes	No	Yes	Yes	No
Observations	1,210	1,210	1,115	1,115	1,115
R-squared	0.400	0.260	0.407		0.226
Number of groups	104	104	104	104	104
F-statistic	15.73***	5.01***	11.98***		3.00***
Wald chi-square statistic				929.76***

Robust standard errors in parentheses.

*** p < 0.01, ** p < 0.05, * p < 0.1

Table 7. Causal mediation analysis output for the study’s outcome model with mediator

	Baseline Model with ROA as outcome variable	Robustness Test Model with Tobins’s Q as outcome variable
	(1)	(2)
Variables
CGQ	0.882**	0.0730*
	(0.401)	(0.375)
FRQ	2.746***	0.0762*
	(0.481)	(0.0450)
SIZE	1.999***	1.461***
	(0.509)	(0.0476)
GRWTH	0.384***	0.154***
	(0.064)	(0.00601)
AGE	0.924*	0.0368
	(0.536)	(0.0502)
LEV	−2.193***	−0.464
	(0.563)	(0.0527)
IFRS	2.710**	0.000207
	(1.267)	(0.119)
ASSTANG	4.096**	0.506***
	(1.640)	(0.154)
Constant	1,930***	106.176***
	(234.81)	(21.990)
Country fixed-effects	Yes	Yes
Year fixed-effects	Yes	Yes
Firm fixed-effects	Yes	No
Observations	1,210	1,210
R-squared	0.139	0.400
Adj. R-squared	0.131	0.396
F-statistic	17.53***	72.60***

Robust standard errors in parentheses.

*** p < 0.01, ** p < 0.05, * p < 0.1

Table 8. Causal mediation analysis output for the study’s mediator model

	Baseline Model with FRQ as outcome variable	Robustness Test Model with FRQ as outcome variable
	(1)	(2)
Variables
CGQ	0.046*	0.0460*
	(0.024)	(0.024)
SIZE	0.006	0.00575
	(0.031)	(0.0306)
GRWTH	−0.003	−0.00339
	(0.004)	(0.00385)
AGE	−0.018	0.0177
	(0.032)	(0.0322)
LEV	0.005	0.00461
	(0.034)	(0.0338)
IFRS	0.032	0.0321
	(0.076)	(0.0761)
ASSTANG	−0.184*	−0.184*
	(0.098)	(0.0984)
Constant	2.820***	2.820
	(14.104)	(14.104)
Country fixed-effects	Yes	Yes
Year fixed-effects	Yes	Yes
Firm fixed-effects	Yes	No
Observations	1,210	1,210
R-squared	0.010	0.010
Adj. R-squared	0.002	0.002
F-statistic	1.24	1.24

Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1

Table 9. The estimates of causal effect of CG on firm performance mediated by FRQ

Variables	Baseline Causal Mediation-Model (1)	Robustness Test Model of Causal Mediation (2)
CDE	0.882**	0.0730*
	(0.401)	(0.0375)
NIE	0.126*	0.00350
	(0.0696)	(0.00276)
TE	1.008**	0.0765**
	(0.405)	(0.0375)
Observations	1,210	1,210

Notes: This table reports the empirical results of controlled effect of CG on firm performance as mediated by EM via means of a causal mediation analysis, where cde=controlled direct effect, nie=natural indirect effect, te=total effect. Asterisks indicate significance at 10% (*), 5% (**) and 1% (***).

Source: Stata output of causal mediation analysis to demonstrate the direct and indirect effects of CG on performance through EM as mediator using the “paramed” command by Emsley and Liu (2013).

Figure 3. Path analysis: Corporate governance, earnings management, and firm performance.

Source: Authors’ framework of mediation analysis involving CG, EM and ROA (2023).

Table

Name of Variable Corporate Governance Mechanisms	Type of Variable	Criteria
Panel A: Board of Directors Board Size Board Independence Board Gender Diversity Board Remuneration Board Meetings Panel B: Audit Audit Committee Size Audit Committee Independence Audit Committee Meeting Audit Committee Diversity Big4 Auditor Engagement Auditor Fee Audit Report Timeliness Panel C: Other Board Committees Remuneration Committee (RC) Size Remuneration Committee Meetings Remuneration Committee Independence Remuneration Committee Gender Diversity Nomination Committee (NC) Size Nomination Committee Meetings Nomination Committee Independence Nomination Committee Gender Diversity Risk/Gov. Committee (RC/GC) Size Risk/Gov. Committee Meetings Risk/Gov. Committee Independence Risk/Gov. Committee Gender Diversity Total Number of Board Committees (including audit committee)	Continuous Ratio Ratio Ratio Continuous Continuous Ratio Continuous Ratio Dummy Ratio Continuous Continuous Continuous Ratio Ratio Continuous Continuous Ratio Ratio Continuous Continuous Ratio Ratio Continuous	Number of members on the board Ratio of non-executive directors to board size Ratio of females to board size Ratio of board wages to total sales Number of board meetings in a year Number of members on Audit Committee (AC) Ratio of outside directors on AC to AC size Number of AC meetings in a year Ratio of female members on AC to AC size If Big4 is engaged 1, otherwise 0 If Big4 is engaged 1, otherwise 0 Days it takes for audit report to be signed and made available to management Number of members on RC Number of RC meetings in a year Ratio of outside directors on RC to RC size Ratio of female members on RC to RC size Number of members on NC Number of NC meetings in a year Ratio of outside directors on NC to NC size Ratio of female members on NC to NC size Number of members on RC/GC Number of RC/GC meetings in a year Ratio of outside directors on RC/GC to its size Ratio of female members on RC/GC to its size Number of board committees functioning on behalf of the board of directors

Source: Authors’ Compilation, 2023.

Pae, J. (2005). Expected accrual models: The impact of operating cash flows and reversals of accruals. Review of Quantitative Finance & Accounting, 24(1), 5–22. https://doi.org/10.1007/s11156-005-5324-7

 Google Scholar

Emsley, R., & Liu, H. (2013). PARAMED: Stata module to perform causal mediation analysis using parametric regression models. In C. Baum. (Ed.), Statistical software components. Boston College Department of Economics. https://econpapers.repec.org/software/bocbocode/S457581.htm

 Google Scholar