Auditor industry specialization and stock price crash risk: individual-level evidence

ABSTRACT

We investigate the effect of individual auditor industry specialization on stock price crash risk. Although research on individual auditors has been growing, we are not aware of any prior studies that investigate industry specialization at the individual auditor-level and crash risk. Using a large sample of Chinese stocks spanning the period 2003–2015, we find a statistically significant and negative association between individual auditor industry specialization and stock price crash risk after controlling the firm-level effect. Our mediation tests suggest that individual auditor industry specialization decreases the risk of price crash by mitigating earnings manipulation. We further document that the negative association is more pronounced for firms that switch from non-specialist to specialist auditors. We also find some evidence that an auditor’s personal characteristics moderate the association between auditor industry specialization and crash risk. Our results remain robust to alternative measures of individual auditor industry specialization and sensitivity checks.

KEYWORDS:

• Auditor industry specialization
• stock price crash risk
• earnings quality
• China

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. For the robustness test on the measures of individual auditor industry expert (EXPERT1 and EXPERT2), we also conduct a test using alternative definitions, which requires both of the auditors to have industry expertise. The result remains consistent with our findings based on the original definition as discussed in the main text.

2. INS HOLD is the sum of Fund Hold Proportion; QFII Hold Proportion; Broker Hold Proportion; Insurance Hold Proportion; Security Fund Hold Proportion; Entrust Hold Proportion; Finance Hold Proportion and Bank Hold Proportion.

3. Calculated as (−0.0701 (coefficient on EXPERT1 in Column (1) *0.43 (S.D. of EXPERT1/0.33 (mean of NCSKEW in t+1).

4. We report the mediation test results for EXPERT1 measure only, to conserve space. Results using EXPERT2 are qualitatively similar to the EXPERT1 results. Results for EXPERT2 are available from the authors upon request.

5. The coefficient on SWITCH_E1(2)_t is −0.0525(−0.0729) and significant at 10% (5%) for the ∆NCSKEW_t+1 measure; the coefficient on SWITCH_E1(2)_t is −0.0263 (−0.0476) and insignificant (significant at 10%) for ∆DUVOL_t+1 measure.

6. The coefficient on SWITCH_NE1(2)_t is 0.0316 (0.0407) and insignificant for the ∆NCSKEW_t+1 measure; the coefficient on SWITCH_NE1(2)_t is 0.0295 (0.0355) and insignificant for the ∆DUVOL_t+1 measure.

7. We acknowledge the circumstances when both of the auditor switch variables SWITCH_E1(2)_t and SWITCH_NE1(2)_t equal zero at the same time, i.e. client firm switches from one expert to another expert, or switches from one non-expert to another non-expert. To accommodate these circumstances, we include both variables SWITCH_E1(2)_t and SWITCH_NE1(2)_t into our regression model. We also conduct regression analysis using each of the two variables separately, the results remain robust (untabulated).

8. Jiang, Wang, and Wang (2018) find that after switching from non-Big N to Big N auditors, the audit quality improved, as measured through discretionary accruals and financial statement divergence scores.

9.

$\begin{array}{rl}& \mathrm{P}\mathrm{r}\left(\mathrm{E}\mathrm{X}\mathrm{P}\mathrm{E}\mathrm{R}{\mathrm{T}}_{\mathrm{i},\mathrm{t}}\right)={\tau }_1\ast \mathrm{S}\mathrm{I}\mathrm{Z}{\mathrm{E}}_{\mathrm{i},\mathrm{t}}+{\tau }_2\ast \mathrm{I}\mathrm{N}{\mathrm{V}}_{\mathrm{i},\mathrm{t}}+{\tau }_3\ast \mathrm{R}\mathrm{E}{\mathrm{C}}_{\mathrm{i},\mathrm{t}}+{\tau }_4\ast \mathrm{I}\mathrm{N}\mathrm{T}\mathrm{A}\mathrm{N}\mathrm{A}\mathrm{S}\mathrm{S}\mathrm{E}{\mathrm{T}}_{\mathrm{i},\mathrm{t}}+{\tau }_5\ast \mathrm{C}{\mathrm{R}}_{\mathrm{i},\mathrm{t}}+\\ & {\tau }_6\ast \mathrm{B}\mathrm{T}{\mathrm{M}}_{\mathrm{i},\mathrm{t}}+{\tau }_7\ast \mathrm{R}\mathrm{O}{\mathrm{A}}_{\mathrm{i},\mathrm{t}}+{\tau }_8\ast \mathrm{L}\mathrm{E}{\mathrm{V}}_{\mathrm{i},\mathrm{t}}+{\tau }_9\ast \mathrm{L}\mathrm{A}\mathrm{G}\mathrm{A}\mathrm{U}\mathrm{O}{\mathrm{P}}_{\mathrm{i},\mathrm{t}}+{\tau }_{10}\ast \mathrm{C}\mathrm{F}\mathrm{F}{\mathrm{O}}_{\mathrm{i},\mathrm{t}}+{\tau }_{11}\ast \mathrm{G}\mathrm{R}\mathrm{O}\mathrm{W}\mathrm{T}{\mathrm{H}}_{\mathrm{i},\mathrm{t}}+{\tau }_{12}\ast \\ & \mathrm{B}\mathrm{O}\mathrm{A}\mathrm{R}\mathrm{D}\mathrm{S}\mathrm{I}\mathrm{Z}{\mathrm{E}}_{\mathrm{i},\mathrm{t}}+{\tau }_{13}\ast \mathrm{I}\mathrm{N}\mathrm{D}\mathrm{E}{\mathrm{P}}_{\mathrm{i},\mathrm{t}}+{\sum \mathrm{Y}\mathrm{E}\mathrm{A}\mathrm{R}}^{+}{\sum \mathrm{I}\mathrm{N}\mathrm{D}}^{+}{\epsilon }_{\mathrm{i},\mathrm{t}}\end{array}$

Where INV is inventories divided by total assets; REC is receivables (net of allowances for bad debts) divided by total assets; INTAN ASSET is intangible assets as a proportion of total assets; CR is current ratio measured as current assets divided by current liabilities; LAG AUOP (Indicator variable equal to one if the firm receives a modified audit opinion in year t-1 and zero otherwise, where a modified opinion is defined as anything other than a standard unqualified audit opinion); CFFO is cash flow from operations divided by total assets; GROWTH is the change of sales between year t and year t-1 divided by total assets; BOARDSIZE is number of board members. Other variables are defined as before.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [71302195,71502134,71672141];