Business Groups and Tunneling: Evidence from Corporate Charitable Contributions by Korean Companies

Abstract

This paper investigates whether corporate philanthropic decisions are associated with a firm’s listing status and business group affiliation. Analyzing a large sample of public and private firms in Korea, we find that (1) public firms make more charitable contributions than private firms and (2) business group-affiliated firms make more charitable contributions than non-affiliated firms. The results suggest that public firms, owing to greater public scrutiny, and business groups, owing to higher political costs, are encouraged to make more corporate charitable contributions. Further, we find that (3) greater corporate giving by public firms than private firms is more pronounced for business group-affiliated firms, compared with non-affiliated firms. The result is consistent with business groups’ strategic coordination of their affiliates’ philanthropic decisions to tunnel business group resources out to controlling shareholders who hold a larger portion of private affiliates than public affiliates.

Appendices

Appendix 1: Variable Descriptions

Variable	Definition
Giving	The amount of charitable contributions as a percentage of sales
GivingD	An indicator variable that equals one if a firm makes charitable contributions
Public	An indicator variable that equals one if a firm is listed on KRX
Affiliate	An indicator variable that equals one if a firm belongs to a business group
FirmSize	The natural log of total assets
Age	The number of years elapsed since the incorporation
Leverage	The book value of short- and long-term debt scaled by the book value of equity
ROA	Net income divided by total assets
Advertising	Advertising expense scaled by total assets
PreGivingD	An indicator variable that equals one if a firm made charitable contributions in the preceding year
Big4Audit	An indicator variable that equals one if a firm has a big 4 auditor
ExportRatio	The ratio of the amount of exports to total sales
QuickRatio	The quick ratio measured by current assets less inventory divided by current liabilities
Industry_PubPr	Proportion of public firms in a firm’s industry
HHI	Herfindahl–Hirschman index, measured as the sum of squares of the market shares (firm sales/industry sales) of the firms in the industry
ΔSALE	Percentage change in sales
ΔGiving	Change in the charitable contributions, scaled by sales
ΔPublicGiving	Change in the sum of the charitable contributions of public affiliates, scaled by the sum of the sales of public affiliates
ΔPrivateGiving	Change in the sum of the charitable contributions of private affiliates, scaled by the sum of the sales of private affiliates
ΔGroup-wide Giving	Change in the sum of the charitable contributions of all the group affiliates, scaled by the sum of the sales of group affiliates
ΔGroup-wide Advertising	Change in the sum of the advertising expense of all the group affiliate, scaled by the sum of the sales of group affiliates
Wedge	The difference between control rights and cash flow rights held by the controlling shareholder
Unrecorded-Marketing Asset	(2.5/3) × Advertisingt + (1.5/3) × Advertisingt−1 + (0.5/3) × Advertisingt−2
CGVD	An indicator variable that equals one if a firm is not classified as a small and medium size enterprise according to Framework Act on Small and Medium Enterprises
G_KFTC	An indicator variable for large business groups or Chaebols as designated by KFTC
KOSPI	An indicator variable for firms on the KOSPI market
KOSDAQ	An indicator variable for firms on the KOSDAQ market
Big30	An indicator variable that equals one if a firm belongs to top 30 Chaebol
Initiation	An indicator variable for the initiation of corporate giving by firms that have not made charitable contributions in the preceding three years

Appendix 2: Description on the Statistical Model

$$Giving_{{i,t}} = {\mkern 1mu} \delta _{0} + {\mkern 1mu} \delta _{1} Public_{{i,t}} + {\mkern 1mu} \delta _{2} Affiliate_{{i,t}} + {\mkern 1mu} \delta _{3} Public_{{i,t}} \times {\mkern 1mu} Affiliate_{{i,t}} + {\mkern 1mu} \delta _{4} FirmSize_{{i,t}} + {\mkern 1mu} \delta _{5} Age_{{i,t}} + {\mkern 1mu} \delta _{6} Leverage_{{i,t}} + {\mkern 1mu} \delta _{7} ROA_{{i,t}} + {\mkern 1mu} \delta _{8} Advertising_{{i,t}} + {\mkern 1mu} \delta _{9} PreGivingD_{{i,t}} + {\mkern 1mu} \delta _{{10}} Big4Audit_{{i,t}} + {\mkern 1mu} \delta _{{11}} IMR{\mkern 1mu} + {\mkern 1mu} \left( {{\text{Year dummies}}} \right){\mkern 1mu} + {\mkern 1mu} \left( {{\text{Industry dummies}}} \right){\mkern 1mu} + \varepsilon,$$

(4)

	Public	Private
Business Group	Public × Affiliate [A]	Private × Affiliate [C]
Non-Group	Public × Non-affiliate [B]	Private × Non-affiliate [D]

1. A: δ₀ + δ₁ + δ₂ + δ₃, B: δ₀ + δ₁, C: δ₀ + δ₂, D:δ₀, A − B = δ₂ + δ₃, A − C = δ₁ + δ₃, B − D  = δ₁, C − D  = δ₂

Equation (4) includes two variables of interest, Public and Affiliate, and an interaction variable of the two. In effect, the sample is partitioned into four subgroups: Public and affiliated firms [A], Public, but not affiliated firms [B], Private but affiliated [C], and Private and not affiliated firms [D]. δ₁ is an estimate of the effect of listing status on corporate charitable contributions and δ₂ is an estimate of the effect of group affiliation on corporate charitable contributions. δ₃ is an estimate of the excessive effect of subgroup A on corporate charitable contributions. The impact of subgroup A (B) on charitable contributions, on average, is computed by summing up δ₀, δ₁, δ_2, and δ₃ (δ₀ and δ₁), while the impact of subgroup C (D) on charitable contributions is sum of δ₀ and δ₂ (δ₀). The different impact on charitable contributions between A and B is the sum of δ₂ and δ₃, while it between A and C is the sum of δ₁ and δ₃. δ₁ (δ₂) shows the different influence on charitable contributions between B and D (C and D).