Earnings management by controlling shareholders who plan for stock gifts: Korean evidence^*

* Accepted by Jeong-Bon Kim upon recommendation by Cheong Yi.

Abstract

We examine whether controlling shareholders who plan for stock gifts would manage earnings in an attempt to depress stock prices prior to gifting stocks to related parties. Gift taxes are levied based on the average market value of the stock transferred for a certain period known as the valuation period. This process enables controlling shareholders to be incentivized to depress stock prices during this period and thereby alleviate tax burden. We discover that the firms that have stock gift transactions in the sample significantly decrease their discretionary accruals in the quarters that precede and/or overlap with the valuation period. Earnings management that decreases income is statistically significant when stock gifts are made for individuals who are the related parties and family members of controlling shareholders. By contrast, we do not observe a similar earnings-management behavior in cases where stock gifts are donated to institutional donees who are not subject to gift taxes.

Keywords:

• Stock gifts
• gift taxes
• earnings management
• controlling shareholders

Acknowledgements

We are grateful to Jeong-Bon Kim (Editor), G. Ryan Huston, Dan Norris, an anonymous reviewer, workshop participants at Seoul National University, and the conference participants at the 2014 AAA annual meeting for their valuable suggestions and comments.

Notes

* Accepted by Jeong-Bon Kim upon recommendation by Cheong Yi.

1. Estate and gift tax rates in Korea are as follows.

2. China Minsheng Banking Corp. and Hurun Research Institute.

3. It is cited from an article in China Youth Daily published on 3 July 2013.

4. A stock gift can be revoked up to three months after it is announced.

5. Unlike in the US, the grantees pay gift taxes in Korea.

6. A bear market occurred in 2000, 2002, and 2008. Given the absence of any standard definition, we define ‘bear market’ as the period where we find consecutive negative returns of KOSPI composite index for three or more quarters.

7. We also use discretionary accruals (DAC) without performance-adjustment as an alternative dependent variable. The results remain qualitatively the same.

8. DAC is estimated cross-sectionally each quarter using all firm-quarter samples in the same one-digit KRX industry code. For industry classification, we follow the KRX sector specification.

9. For the period after 2003, total accruals are measured as the difference between net income and cash flow from operations. For the period before 2003, total accruals are measured using an indirect approach because the data on cash flow from operations in Korea are not available for most of the sample.

where TA_it = total accruals, defined as ∆current assets − ∆current liabilities − ∆cash + ∆short-term debt − depreciation for firm i in quarter t.

10. We include a dummy variable for the fourth fiscal quarter because accruals in the fourth quarter may differ from accruals in the first three quarters due to increased auditor inspection and discretionary asset write-offs (Francis, Hanna, and Vincent 1996; Matsumoto 2002).

11. Listed firms in Korea are required by the Securities and Exchange Act to release their quarterly earnings within 45 days after a quarter ends. Further, Korean listed firms rarely report quarterly earnings within 30 days.

12. We have also used discretionary accruals (DAC) without performance adjustment as an alternative dependent variable. However, the results remain qualitatively the same.

13. Variables for determination model include largest shareholder ownership, size, asset growth, cash flow from operations (CFO), lagged accruals, and lagged loss indicator.

14. One to one matching with no replacement. The Caliper width used for matching is 0.05sigma. Control sample for PSM test is constructed from the listed firms that do not involve any gift transactions throughout the sample period.