Determinants of industry expertise outsourced IAF: Do company and auditor attributes affect the selection?

Figures & data

Table 1. Variables Definition

Variable	Definitions
	Dependent variables	Data sources
INDSOIAF30	The indicator variable equals 1 if the external IAF provider is an industry specialist who shares 30 percent of market share, 0 otherwise.	CG report & DataStream
INDSOIAF20	The indicator variable equals 1 if the external IAF provider is an industry specialist who shares 20 percent of market share, 0 otherwise.	CG report & DataStream
	Independent variables
	Company characteristics
LNSALE	The natural log of total sales.	DataStream
MB	The ratio of equity market to equity book value.	DataStream
LEV	The total debt divided by total assets.	DataStream
LOSS	An indicator variable equals 1 if the company current performance is loss, 0 otherwise.	DataStream
QUICK	The ratio of current assets minus inventory divided by current liabilities.	DataStream
ATURN	The ratio of current sales/revenues divided by total assets.	DataStream
OWCCO	The percentage of common shares held by major shareholders (≤10%).	CG report
NWEQTY	An indicator variable equals 1 if the company issues new common shares in the year, 0 otherwise.	Financial Reports
LNAG	The natural log of the number of years since the company establishment.	Financial Reports
TACC	The proportion of net income minus operating cash flow divided by total assets	DataStream
	External auditor characteristics
ADFSIZE	An indicator variable equals 1 if the company is audited by one of big4 audit firms, 0 otherwise.	Audit report
ADFIND	The indicator variable equals 1 if the company is audited by industry-specialist audit firm, 0 otherwise.	CG report & Audit report
LNADFEE	The natural log of audit fees paid to external auditor for statutory audit.	CG report
	Control variables
YFIX	The year indicators for the years from 2010 to 2017.	DataStream
INDFIX	The industry indicators for the industrial, consumer discretionary, materials, and consumer staples.	DataStream

Table 2. Sample distribution

Panel A: Sample size	Obs.
Number of observations for listed companies over the period 2010–2017	935
Less: Observations for financial companies	(271)
Observations for companies with in-house IAF	(225)
Observations with missing data	(105)
Final observations for testing the hypotheses	334
Panel B: Sample distribution by industry
Industry	Obs.
Manufacturing	30
Energy	43
Consumer discretionary	62
Materials	85
Consumer staples	114
Obs.	334
Industrial	225
Services	109
Obs.	334

Table 3. Descriptive statistics

Variable	Obs	Mean	Std. Dev.	Min	Max	INDSOIAF	NINDSOIAF	DIFF.
INDSOIAF30	334	0.243	0.429	0.000	1.000	-	-	-
INDSOIAF20	334	0.305	0.461	0.000	1.000	-	-	-
LNSALE	334	9.065	1.377	0.000	11.604	9.945	8.783	1.162***
MB	334	23.963	409.553	−5.380	7486.34	2.056	1.423	0.633**
LEV	334	0.466	0.280	0.013	1.769	0.483	0.460	0.022
LOSS	334	0.159	0.366	0.000	1.000	0.086	0.182	−0.095**
QUICK	334	1.901	3.845	0.039	62.504	1.291	1.895	−0.604***
ATURN	334	0.788	0.475	0.000	2.538	0.873	0.760	0.112
OWCCO	334	60.556	22.381	13.100	99.130	55.913	62.043	−6.130**
NWEQTY	334	0.138	0.345	0.000	1.000	0.074	0.158	−0.084**
LNAG	334	2.999	0.493	0.000	3.761	3.034	2.998	0.037
TACC	334	−0.043	0.126	−0.829	0.851	−0.057	−0.038	0.019
ADFSIZE	334	0.617	0.487	0.000	1.000	0.765	0.569	0.196***
ADFIND	334	0.320	0.467	0.000	1.000	0.358	0.308	0.050
LNADFEE	334	8.754	0.495	7.783	10.219	9.031	8.666	0.365***

*** p < 0.01; ** p < 0.05; * p < 0.10;

INDSOIAF is an indicator for outsourced IAF industry specialist sample; NINDSOIAF is an indicator for outsourced IAF non-industry specialist sample; See Table 1 for variables definition.

Table 4. Correlation matrix based on Pearson analysis

Variables	−1	−2	−3	−4	−5	−6	−7	−8	−9	−10	−11	−12	−13	−14	−15
(1) INDSOIAF30	1
(2) INDSOIAF20	0.85*	1
(3) LNSALE	0.36*	0.42*	1
(4) LEV	0.03	0.05	0.15*	1
(5) OWCCO	−0.12*	−0.13*	−0.36*	0.04	1
(6) NWEQTY	−0.11*	−0.11*	−0.04	−0.04	−0.02	1
(7) MB	0.17*	0.16*	0.24*	0.04	−0.06	−0.05	1
(8) LOSS	−0.11*	−0.15*	−0.32*	0.36*	0.15*	−0.06	−0.15*	1
(9) QUICK	−0.13*	−0.13*	−0.29*	−0.46*	−0.16*	−0.04	0.12*	−0.06	1
(10) ATURN	0.10*	0.15*	0.33*	0.05	−0.35*	0	−0.02	−0.06	−0.04	1
(11) TACC	−0.08	−0.07	0.16*	−0.05	−0.11*	−0.07	0	−0.21*	−0.06	0.01	1
(12) LNAG	0.03	0.08	−0.05	−0.25*	−0.18*	−0.07	−0.38*	0.06	0	0.36*	0.03	1
(13) ADFSIZE	0.17*	0.16*	0.34*	0.11*	−0.20*	0.10*	0.16*	−0.16*	−0.05	−0.07	0.09	−0.19*	1
(14) ADFIND	0.05	0.07	0.17*	−0.01	0.04	0.01	0.02	−0.05	−0.12*	−0.15*	0.01	0.08	0.28*	1
(15) LNADFEE	0.32*	0.33*	0.67*	0.12*	−0.20*	0.01	0.29*	−0.27*	−0.16*	−0.08	0.09	−0.17*	0.55*	0.27*	1

*p < 0.05; See Table 1 for variables definition.

Table 5. Logistic regression results for testing the hypotheses

Variable		(1)		(2)
	Sign	INDOIAF30		INDOIAF20
		Estimate	T.stat	Estimate	T.stat
LNSALE	+	1.278***	(4.227)	1.811***	(5.448)
MB	+	0.169*	(1.805)	0.238**	(2.282)
LEV	+	−0.435	(−0.394)	1.002	(0.986)
LOSS	+	0.145	(0.234)	−0.113	(−0.175)
QUICK	-	−0.112	(−0.971)	0.056	(0.512)
ATURN	-	−0.202	(−0.432)	−0.416	(−1.016)
OWCCO	-	−0.003	(−0.298)	0.008	(0.958)
NWEQTY	+	−1.315**	(−1.976)	−1.526**	(−2.093)
LNAG	-	1.366**	(2.058)	2.619***	(3.996)
TACC	+	−6.024***	(−2.692)	−6.223***	(−3.197)
ADFSIZE	+	0.297	(0.618)	0.036	(0.079)
ADFIND	+	0.137	(0.316)	0.125	(0.307)
LNADFEE	+	−0.232	(−0.398)	−0.745	(−1.334)
YFIX	?	CONTROLLED
INDFIX	?	CONTROLLED
_cons	?	−13.865***	(−3.322)	−18.352***	(−4.010)
Observations		334	334
Pseudo R^2		29%	34%
Wald chi-square	66.57***	72.83***

*** p < 0.01, ** p < 0.05, * p < 0.10

See Table 1 for definitions.

Table 6. Logistic regression results for the types of INDOIAF30

Variable	(1)		(2)
	BIG4		SECNDT
	Estimate	T.stat	Estimate	T.stat
LNSALE	3.125***	(3.159)	0.846	(1.307)
MB	−1.824*	(−1.957)	0.155	(0.811)
LEV	−7.053**	(−2.573)	2.706*	(1.771)
LOSS	−1.771	(−0.959)	−2.458**	(−2.508)
QUICK	−1.363**	(−2.234)	−0.076	(−0.327)
ATURN	−1.937	(−1.562)	1.779**	(2.400)
OWCCO	−0.059	(−1.399)	0.029**	(2.075)
NWEQTY	−0.934	(−0.596)	0.702	(0.736)
LNAG	2.442	(0.804)	3.011**	(2.343)
TACC	3.091	(0.666)	−10.825***	(−3.490)
ADFSIZE	2.325	(1.512)	1.742**	(2.320)
ADFIND	0.493	(0.631)	0.284	(0.308)
LNADFEE	−3.114	(−0.827)	−2.367**	(−2.034)
YFIX	CONTROLLED
INDFIX	CONTROLLED
_cons	−22.836	(−0.686)	−21.743***	(−3.829)
Observations	334		334
Pseudo R^2	57%		42%
Wald chi-square	221.70***		578.56***

*** p < 0.01, ** p < 0.05, * p < 0.10

BIG4 is an indicator for the analysis of INDOIAF30 if the type of provider is a big4 audit firm; SECNDT is an indicator for the analysis of INDOIAF30 if the type of provider is a second-tier audit firm; See Table 1 for definitions.