Internal Capital Markets in Business Groups: Evidence from the Asian Financial Crisis

• DOI: http://doi.org/10.1111/jofi.12309
• Date: 01 December 2015
• Author: HEITOR ALMEIDA,CHANG‐SOO KIM,HWANKI BRIAN KIM
• Published date: 01 December 2015

THE JOURNAL OF FINANCE •VOL. LXX, NO. 6 •DECEMBER 2015

Internal Capital Markets in Business Groups:

Evidence from the Asian Financial Crisis

HEITOR ALMEIDA, CHANG-SOO KIM, and HWANKI BRIAN KIM∗

ABSTRACT

This paper examines capital reallocation among firms in Korean business groups

(chaebol) in the aftermath of the 1997 Asian financial crisis, and the consequences of

this capital reallocation for the investment and performance of chaebol firms. Weshow

that chaebol transferred cash from low-growth to high-growth member firms, using

cross-firm equity investments. This capital reallocation allowed chaebol firms with

greater investment opportunities to invest more than control firms after the crisis.

These firms also showed higher profitability and lower declines in valuation than

control firms following the crisis. Our results suggest that chaebol internal capital

markets helped them mitigate the negative effects of the Asian crisis on investment

and performance.

The literature on internal capital markets focuses mostly on multisegment

firms (conglomerates), which are common in the United States. A “business

group”—a group of legally independent firms under common ownership—

shares many of the features of U.S.-type conglomerates. In particular, both

have internal markets that can allocate capital among member firms and divi-

sions. While the importance of internal capital allocation within conglomerates

has probably decreased in the United States as a result of financial develop-

ment, the role of groups may still be significant in countries with weak institu-

tions and less developed financial markets (Khanna and Palepu (2000), Morck,

Wolfenzon, and Yeung (2005), Khanna and Yafeh (2007)). Notwithstanding the

potential importance and ubiquity of groups in emerging markets (Claessens

et al. (2002), Faccio and Lang (2002)), however, there is surprisingly little direct

evidence on the efficiency of their capital allocation.

In this paper,we study capital allocation in Korean business groups (chaebol),

focusing on an event that is likely to have exacerbated the impact of chaebol

∗Chang-Soo Kim (corresponding author) is at YonseiUniversity. Hwanki Brian Kim and Heitor

Almeida are at the University of Illinois at Urbana–Champaign. We thank three anonymous

referees, Michael Roberts, Ran Duchin, Miguel Ferreira, Antonio Gledson de Carvalho, Radha

Gopalan, David Reeb, Mara Faccio, Patrick Behr, and participants at Hong Kong University, Pur-

due University, Bocconi University, the 2012 SBFIN conference, the 2012 LUBRAFIN conference,

the Washington University 2011 Corporate Finance Conference, and the HKUST 2011 Finance

Symposium for comments and suggestions. We also thank Igor Cunha, Jiyoon Lee, Yunhee Cho,

Sunghoon Joo, and Myeongwoo Kang for excellent research assistance. The authors do not have

any conflicts of interest, as identified in the Disclose Policy.

DOI: 10.1111/jofi.12309

2539

2540 The Journal of Finance R

internal capital markets on resource allocation, namely,the Asian crisis of 1997.

The 1997 Asian crisis was a truly unprecedented and unexpected event that

greatly affected Asian economies. Important sources of external finance (such

as bank loans and bonds) became more costly or virtually impossible to access

in the immediate aftermath of the crisis. In such an environment of financial

stress, the ability of chaebol to reallocate resources across member firms may

have been particularly important. On the other hand, the crisis may have

exacerbated managers’ incentives to engage in tunneling and expropriation

due to its impact on the expected rate of return on investment (Johnson et al.

(2000)). The aftermath of the Asian financial crisis is thus a good laboratory to

study both the bright and the dark sides of business groups’ internal capital

markets.

We start by establishing direct evidence that chaebol increased capital reallo-

cation across member firms in the aftermath of the crisis. Using hand-collected

data on equity investments by chaebol members, we show that, in the two

years following the crisis (i.e., 1998 to 1999), chaebol firms with the highest

growth opportunities sold equity to other member firms in exchange for cash.

This direction of capital reallocation is consistent with efficient internal capital

markets. The magnitude of these intrachaebol capital transfers is also substan-

tial. For example, aggregate equity transfers increased from approximately 1%

of total chaebol assets in the precrisis period (1996 to 1997) to 2.5% of chaebol

assets in the postcrisis period (1998 to 1999). A one-standard-deviation change

in our preferred proxy for growth opportunities (Tobin’s Qmeasured at the

industry level) is associated with an increase in postcrisis equity transfers of

2.6% of firm assets. Thus, chaebol firms with favorable growth prospects re-

ceived economically significant transfers of capital following the Asian crisis.1

Notably,this reallocation pattern does not hold in the precrisis period, suggest-

ing that the reallocation was indeed associated with the Asian crisis.

Next, we examine the consequences of this capital reallocation for the in-

vestment and performance of chaebol firms. In short, our empirical strategy

is to compare changes in investment and other outcomes across chaebol and

nonchaebol control firms from the precrisis period to the postcrisis period. Since

the Asian crisis was unexpected, it is unlikely that chaebol changed their struc-

ture in anticipation of the crisis. We select the nonchaebol control firms via a

matching procedure to minimize the impact of observable precrisis differences

between chaebol and control firms. We also conduct a battery of placebo and

other tests to rule out alternative explanations that rely on endogenous se-

lection of firms into chaebol as well as demand effects that may differentially

affect chaebol and control firms following the crisis.

Our results suggest that chaebol firms were able to invest more on aver-

age than control firms in the aftermath of the crisis. In particular, while con-

trol firms are indistinguishable from chaebol firms along several (precrisis)

1We also find some evidence that chaebol firms with lower liquidity received equity transfers

after the crisis, though this evidence holds only for an industry-level proxy for external finance

dependence (Rajan and Zingales (1998)).

Internal Capital Markets in Business Groups 2541

dimensions, including Tobin’s Q, profitability,cash holdings, investment, lagged

investment growth, and leverage, their annual investment declines from 10.6%

of assets precrisis to 2.1% of assets postcrisis, compared to a decrease from

10.4% of assets precrisis to 6.7% of assets postcrisis for chaebol firms. Both the

difference-in-differences in investment rates across chaebol and control firms

(4.9%) and the average treatment effect on the treated firms estimated by the

matching procedure (4.0%) are economically large and statistically significant.

We do not find differences in investment activities across chaebol and control

firms over normal (i.e., noncrisis) periods, or in a recessionary period that likely

generated a decrease in demand for investment but was not associated with

financial turmoil (1992 to 1993).

We attempt to rule out two alternative explanations for our results. First,

chaebol may have been less affected by increased government regulation fol-

lowing the crisis due to their political links and important status in the Korean

economy. For example, the Korean government encouraged firms to engage in

debt reduction following the crisis. However, we find no significant differences

in proxies for growth in external equity and debt across chaebol and nonchaebol

firms. Second, we show that the results continue to hold for a sample of low

leverage firms, thus providing evidence against the hypothesis that an increase

in distress risk induced by the crisis may explain the differential investment

response across groups.

As we discuss in Section I, the differential investment growth that we observe

can be explained by chaebols’ internal capital markets. The key condition that

generates higher average chaebol investment (relative to nonchaebol firms) is

that some firms have financing capacity greater than desired investment in the

aftermath of the Asian crisis. The crisis represents a shock to both financing ca-

pacity and investment opportunities. Since desired investment likely decreased

after the crisis, some firms may have ended up with financing capacity in excess

of optimal investment. To the extent that some chaebol firms reallocated excess

funds to other chaebol firms that needed liquidity, a reallocation that cannot

happen in the external market, chaebol investment is likely to have fallen less

than investment by similar nonchaebol firms. In line with this view, we find

that the lower decline in investment in chaebol firms is driven by chaebol with

high liquidity, which may have been in a better position to reallocate funds

across firms.

To provide additional evidence on the consequences of internal capital alloca-

tion by chaebol, we examine the distribution of investment across chaebol and

control firms, as well as their relative performance, following the Asian crisis.

Our evidence suggests that the relationship between investment and growth

opportunities in the aftermath of the crisis was stronger among chaebol than

among control firms. Chaebol firms with industry-level Tobin’s Qabove the

chaebol median (our proxy for growth opportunities) do not show significant

declines in investment following the Asian crisis (the average postcrisis change

in investment is −0.06% of assets for this group). In contrast, we observe a large

decline in investment for chaebol firms with low growth opportunities (−5.7%).

In the sample of control firms, the decline in investment is large and negative,