Is the nexus between capital structure and firm performance asymmetric? An emerging market perspective

Figures & data

Table 1. Construct description.

Construct	Operational definition
Regressand: Firm Performance
ROA	Profit After Tax to Total Asset
ROE	Profit After Tax to Total Equity
Regressor: Capital Structure
DA	Total Debt to Total Asset
SDA	Total Short-term Debt to Total Equity
LDA	Total Long-term Debt to Total Asset
Control Variables
Siz	Ln of Total Assets
Grw	$\frac{\text{Current year sales}-\text{Previous year sales}}{\text{Previous year sales}}$
Tang	Total Fixed Asset to Total Asset
Inv to Asst	Total Inventory to Total Current Asset
GDP	Percentage change in annual GDP rate
Infl	Percentage consumer price index
HHI	${\left(\frac{\mathrm{\text{Fir}}{\mathrm{m}}^{\prime }\text{s sales in year t}}{\text{Total sales of industry in year t}}\right)}^2$

Table 2. Descriptive statistics.

Variable	Mean	Median	S.D	Min	Max	VIF results
ROA	0.0891	0.0824	0.1218	−1.5255	0.5161
ROE	0.1795	0.1563	0.2185	−1.5238	1.4788
SDA	0.0639	0.0401	0.0762	0.0005	0.7079	1.252
LDA	0.1163	0.0584	0.1353	0.0006	0.6266	1.757
TDA	0.1651	0.1004	0.1711	0.0008	0.7485	2.172
Siz	12.1003	12.0000	1.5079	5.3968	16.0898	1.316
Grw	0.0808	0.0885	0.3057	−4.7244	0.9942	1.050
Tang	0.2598	0.2369	0.1693	0.0005	0.7404	1.318
Inv to asst	0.1149	0.0885	0.1074	0.0003	0.6847	1.382
GDP	0.5827	0.6800	0.3895	−0.5800	0.9100	1.207
Infl	0.5500	0.5000	0.1547	0.3600	0.9400	1.548
HHI	−0.0663	−0.0188	0.1469	0.0001	1.7917	1.062

Source: Author calculation.

Table 3. Correlation analysis.

Variables	ROA	ROE	SDA	LDA	TDA	Siz	Grw	Tang	Inv to asst	GDP	Infl	HHI
ROA	1
ROE	0.388***	1
SDA	−0.097**	−0.009*	1
LDA	−0.446***	−0.223***	0.366***	1
TDA	−0.451***	−0.317***	0.205***	0.098***	1
Siz	−0.331***	−0.288***	0.189***	0.317***	0.274***	1
Grw	0.081*	0.103**	0.054*	0.014	0.041	0.054	1
Tang	0.148***	0.06	0.366***	0.439***	0.343***	0.133***	0.056	1
Inv to asst	0.184***	0.172***	0.253***	0.247***	0.110*	0.427****	0.106*	0.112*	1
GDP	0.001**	0.006*	0.051	0.042	0.038	0.014	0.117***	0.007	0.012*	1
Infl	−0.074*	−0.098**	−0.054	−0.081**	−0.096**	0.001	0.078*	0.064	0.002	0.295***	1
HHI	0.051**	0.023	0.123***	0.138***	0.095**	0.138***	0.119***	0.171***	0.038	0.001	0.028	1

Source: Author calculation.

***, **, * signifies the level of significance at 1%, 5%, and 10% level.

Table 4. Unit root test results.

Variable	Statistic	Acceptance/rejection of the null hypothesis
ROA	−8.422 ***	Rejected
ROE	−9.102 ***	Rejected
SDA	−33.158 ***	Rejected
LDA	−7.151 ***	Rejected
TDA	−13.399 ***	Rejected
Siz	−5.014 ***	Rejected
Grw	−16.038 ***	Rejected
Tang	−8.294 ***	Rejected
Inv to asset	−8.731 ***	Rejected
GDP	−13.776 ***	Rejected
Infl	−6.934 ***	Rejected
HHI	−5.534 ***	Rejected

${\mathit{H}}_0:$ The data is non-stationary.

***, **, * signifies the level of significance at 1%, 5%, and 10% level.

Table 5. Two-step system GMM regression results.

	Model 1 Dependent: ROA	Model 2 Dependent: ROE
SDA	−0.2219 (0.0122) ***	−0.0861 (0.0037) ***
LDA	−0.7189 (0.0725) ***	−0.1778 (0.0183) ***
TDA	−0.4823 (0.0384) ***	−0.0023 (0.0009) ***
Siz	−0.0004 (0.0007) ***	−0.0029 (0.0016) *
Grw	0.0275 (0.0026) ***	0.0572 (0.0067) ***
Tang	0.0071 (0.0035) **	0.0285 (0.0168) *
Inv to Asst	0.0781 (0.0232) ***	0.4413 (0.0503) ***
GDP	0.0008 (0.0004) **	0.0111 (0.0025) ***
Infl	−0.0044 (0.0024) *	−0.0064 (0.0030) **
HHI	0.0792 (0.0093) ***	0.4019 (0.0263) ***
Constant	0.1372 (0.0083) ***	0.0852 (0.0241) ***
L.ROA	0.5089 (0.0071) ***
L.ROE		0.6289 (0.0053) ***
Firm Year Observations	515	513
AR (1) [p-value]	0.000	0.005
AR (2) [p-value]	0.697	0.862
Hansen J [p-value]	0.347	0.595

Source: Author calculation.

Note: Standard errors are parentheses, ***p < 1%, **p < 5%, *p < 10%. The Hansen J test reports the p-value for the validity of instrumental variables (${\mathit{H}}_0:$ The instrumental variables are valid). The AR (1) and AR (2) report the p-values for the absence of first-order serial correlation and second-order serial correlation of the residuals, respectively (${\mathit{H}}_0:$ The residuals have no first-order/second-order autocorrelation).

Table 6. Quantile regression results.

	Dependent: ROA					Dependent: ROE
Variables	Q(10)	Q(25)	Q(50)	Q(75)	Q(90)	Q(10)	Q(25)	Q(50)	Q(75)	Q(90)
SDA	−0.0174 (0.0082) **	−0.0405 (0.0234) *	−0.0463 (0.0258) *	−0.0589 (0.0206) ***	−0.0645 (0.0183) ***	−0.0014 (0.0007) *	−0.0574 (0.0313) *	−0.0785 (0.0375) **	−0.0874 (0.0379) **	−0.0997 (0.0372) ***
LDA	−0.2804 (0.1285) **	−0.3726 (0.1124) ***	−0.4172 (0.0635) ***	−0.4360 (0.0645) ***	−0.6759 (0.1328) ***	−0.5997 (0.1649) ***	−0.7166 (0.1154) ***	−0.7206 (0.2361) ***	−0.8488 (0.2180) ***	−1.3413 (0.4427) ***
TDA	−0.3668 (0.0935) ***	−0.4503 (0.0861) ***	−0.5246 (0.0503) ***	−0.5391 (0.0414) ***	−0.6568 (0.1084) ***	−0.5723 (0.1151) ***	−0.6594 (0.0791) ***	−0.7198 (0.1634) ***	−0.8121 (0.1095) ***	−1.2607 (0.3679) ***
Siz	−0.0078 (0.0035) **	−0.0079 (0.0017) ***	−0.0083 (0.0028) ***	−0.0118 (0.0034) ***	−0.0126 (0.0021) ***	−0.0105 (0.0042) **	−0.0137 (0.0069) **	−0.0152 (0.0097)	−0.0264 (0.0045) ***	−0.0481 (0.0110) ***
Grw	0.0267 (0.0109) **	0.0298 (0.0147) *	0.0311 (0.0096) ***	0.0338 (0.0412)	0.0458 (0.0169) ***	0.0298 (0.0175) *	0.0449 (0.0281)	0.0468 (0.0261) *	0.0497 (0.0238) **	0.0937 (0.0255) ***
Tang	0.0283 (0.0153) *	0.0333 (0.0147) **	0.0435 (0.0277)	0.0805 (0.0191) ***	0.0876 (0.0353) **	0.0057 (0.0349)	0.0664 (0.0312) **	0.0860 (0.0379) **	0.1097 (0.0500) **	0.1340 (0.0515) ***
Inv to Asst	0.0535 (0.0312) *	0.0757 (0.0572)	0.0830 (0.0272) ***	0.1018 (0.0428) **	0.1751 (0.0411) ***	0.1163 (0.0892)	0.2285 (0.0551) ***	0.2932 (0.0906) ***	0.3631 (0.0808) ***	0.3698 (0.1154) ***
GDP	0.0004 (0.0002) *	0.0033 (0.0017) *	0.0043 (0.0129)	0.0074 (0.0088)	0.0098 (0.0039) ***	0.0036 (0.0017) **	0.0064 (0.0029) **	0.0071 (0.0129)	0.0121 (0.0092)	0.0662 (0.0306) **
Infl	−0.0019 (0.0011) *	−0.0201 (0.0288)	−0.0554 (0.0297) *	−0.0625 (0.0380)	−0.1200 (0.0442) ***	−0.0087 (0.0046) *	−0.0137 (0.0612)	−0.0282 (0.0599)	−0.1227 (0.0664) *	−0.3990 (0.1288) ***
HHI	0.0049 (0.0142)	0.0052 (0.0028) *	0.0151 (0.0079) *	0.0172 (0.0192)	0.0465 (0.0149) ***	0.0021 (0.0014) **	0.0174 (0.0099) *	0.0277 (0.0269)	0.0387 (0.0343)	0.3525 (0.1678) **
Constant	0.0512 (0.0243) **	0.1413 (0.0457) ***	0.1774 (0.0324) ***	0.2638 (0.0318) ***	0.2917 (0.0504) ***	0.1837 (0.1021) *	0.1926 (0.0984) *	0.2325 (0.0709) ***	0.5054 (0.0741) ***	0.7411 (0.1336) ***
Quasi-LR statistic	152.58 ***	161.09 ***	222.38 ***	266.96 ***	233.41 ***	120.39 ***	98.24 ***	110.33 ***	141.96 ***	101.74 ***
Observ (N)	521	521	521	521	521	519	519	519	519	519

Source: Author calculation.

Note: Standard errors are parentheses, ***p < 1%, **p < 5%, *p < 10%. The quantiles, Q(10), Q(25), Q(50), Q(75), Q(90) represent the distribution of firms with the 10^th, 25^th, 50^th, 75^th, and 90^th percentile of firms in terms of profitability.

Table 7. Wald test results.

	Dependent: ROA				Dependent: ROE
Variables	H0: Q(10) =Q(90)	H0: Q(25) =Q(90)	H0: Q(50) =Q(90)	H0: Q(75) =Q(90)	H0: Q(10) =Q(90)	H0: Q(25) =Q(90)	H0: Q(50) =Q(90)	H0: Q(75) =Q(90)
Wald test statistic	88.5074	58.8876	84.0075	20.7323	24.2211	28.7376	36.2985	20.7729
p-value	0.0000	0.0000	0.0000	0.0230	0.0026	0.0014	0.0001	0.0227

Source: Author calculation.

Note: The Wald test examines if the quantile results are constant across different points of the conditional distribution of the dependent variable, i.e. firm performance (${\mathit{H}}_0:$ There is slope equality, and coefficients are equal across different quantiles). ***, **, * signifies the level of significance at 1%, 5%, and 10% level.