Changes in chaebol firms’ overinvestment after the Asian financial crisis: a long-term perspective

Abstract

This paper examines chaebol firms’ overinvestment practices before and after a financial crisis in Korea. In comparison to non-chaebol firms, chaebol firms significantly reduced overinvestment after the Asian financial crisis. Also, chaebol firms were less inclined to engage in earnings management for overinvestment after the financial crisis. The results suggest that governance and financial reforms focusing on chaebol firms and emphasizing long-term firm value after the crisis discouraged chaebol firms’ overinvestment practices, bringing about a sharp rebound in their business performance.

Keywords:

• chaebol
• overinvestment
• earnings management
• Asian financial crisis
• future operating performance

JEL Classifications:

• G18
• G31
• G38
• M40
• M41

Notes

1. Every year, the Korea Fair Trade Commission (KFTC) announces a list of large business groups whose assets are worth more than KRW 2 trillion. The firms included in the list are subject to greater anti-trust scrutiny in their intra-group transactions and take more responsibility in terms of disclosure requirements (Choi, Jen, and Shin 2000).

2. Overinvestment is measured using Tobin’s Q and the McNichols and Stubben (2008) model. The method for estimating overinvestment measured from the two models in Equations (2(2) ) and (3(3) ) is discussed in Section 3.

3. Cash outflows from investing activities are calculated by adding all cash outflow incurred for property, plant, and equipment (McNichols and Stubben 2008; Choi and Kwak 2010).

4. There may be some survival bias in the sample selection process. For example, a firm delisted in 2005 may be excluded from the sample because the firm is not listed on the KSE in 2007. However, the firm should be included in the sample from 1988 to 2004, because it is listed during that period. To control for any sample selection bias that can arise from survivorship, data on firms listed on the KSE are downloaded for each year during the sample period from 1988 to 2007.

5. This study covers observations up to 2007 to avoid the effect of the global financial crisis in 2008.

6. Each year until 2001, the KFTC announced the 30 largest chaebol groups based on total assets. Since 2002, the KFTC has provided a list of chaebol groups whose assets exceed KRW 2 trillion.

7. Figure 3 presents a time series trend in earnings management based on the DR measure. It shows that chaebol firms are generally not more likely to be involved in earnings management than non-chaebol firms. This is inconsistent with the findings of Kim and Yi (2006), who use absolute discretionary accruals instead of DRs. Kim and Yi (2006) examine earnings management by using total accruals because they measure absolute discretionary accruals from the modified Jones (1991) model. On the other hand, this paper investigates how firms manage earnings mainly by using sales and accounts receivable, as in McNichols and Stubben (2008). The dependent variable in this paper is the level of investment, and therefore discretionary accruals from the modified Jones (1991) model may show a mechanical relationship between PPE and the investment level. In addition, firm size and leverage are included in the DR model because chaebol firms are larger and more levered than non-chaebol ones. However, it is acknowledged that the results in this paper may be influenced by measurement errors in overinvestment and underinvestment, and earnings management based on DRs.

8. Although a 2SLS regression analysis is conducted to control for the endogeneity problem between earnings management and overinvestment, endogeneity may still affect results because all explanatory variables in the first stage of the 2SLS regression analysis are endogenous.

9. The second hypothesis addresses the relationships between three endogenous variables, namely chaebol affiliation, overinvestment, and earnings management, and therefore the potential issue of endogeneity cannot be ignored.