Article,
Determinants of family ownership: The choice between control and performance
Asia-Pacific Journal of Financial Studies, 35 (6): 39-75 (2006)
Abstract
Family ownership is a common form of ownership around the world. Numerous
studies have documented family ownership in different economies,
and they show that family ownership and control is not confined to
privately held firms; it is also dominant among publicly traded firms.
These studies also report that controlling families have power over
their firms significantly in excess of their cash flow rights; they
maintain this controlling power through pyramidal ownership structures
and cross-holdings among affiliated companies. Existing literature
has analyzed the relationship between family ownership and firm performance.
These studies do not, however, analyze factors that determine the
family's ownership. In a family controlled firm, the controlling
shareholder and his family would value its control as much as it
would value the firm's performance, if not more. The pyramidal ownership
structure shows that the family really cares about securing the control
on its groups. We hypothesize that controlling families are as much
interested in their controlling power as they are interested in firm
performance. Firm performance is important, but controlling families
may forgo the performance, and subsequent economic benefit therein,
for the sake of securing the control over the firm if their private
benefit from having the control exceeds economic rewards from good
performance of their firms. If the family must choose between performance
and control, it is highly likely to choose control over performance
in many emerging-market countries where the monetary and non-monetary
benefits of having the control may far exceed cash rewards from good
firm performance. Whether controlling families value their controlling
power as much as monetary rewards - either in cash dividends or in
capital gains - from the firm's performance is what this paper sets
out to analyze. It is the purpose of this paper to investigate the
factors that render a controlling Korean family to own shares of
some companies on its own while it lets affiliated companies own
the shares of other companies. This paper also explores the question
of what makes the family and the affiliated companies own more or
fewer shares among these companies. We analyze non-financial firms
listed on the Korea Stock Exchange as of December 31 of each year
for the 1998 ∼ 2001 period. We chose this particular period because
during this period the ceiling on the amount of equity investment
that can be invested in other affiliated companies' equity was removed.
Our period selection is justified by this deregulation that firms
are allowed to make equity investment without limit into affiliated
companies. To maintain consistency of ownership data that is constructed
as of the end of each calendar year, we restrict our sample to firms
whose fiscal year ends on December 31. We also exclude from our sample
those firms which have missing financial data. We identify for each
firm a controlling shareholder who is related to the founding family
regardless of the number of shares the controlling shareholder owns.
If there is no founding family, we identify an individual shareholder
and those family members who own a substantial fraction of outstanding
shares and have effective control of the firm. Firms with no controlling
family are excluded from our sample because our analysis focuses
on what determines family ownership. The final sample selection filter
is the existence of affiliated companies. If a firm does not have
an affiliated company and stands alone as a business entity, all
controlling power should be secured by family ownership. These firms
are excluded from our sample. The application of the above selection
criteria stated above leaves 1,530 firms for the sample period. Our
analysis shows that control is a more important factor than firm
value in the determination of whether an entrepreneurial family in
Korea chooses to own a firm and how much to own. Controlling families
prefer to own the shares of de facto holding companies that hold
relatively larger ownerships of other affilia ed firms in the group
because they provide the controlling families with more control over
affiliated companies. On the other hand, controlling families force
their affiliated companies to own more shares of firms that provide
the families with less of controlling power over other member firms.
To control for the endogeneity problem between family ownership and
other variables such as firm value, affiliated ownership and debt
ratio, we also set up simultaneous models where two-stage least square
method is used to estimate the coefficients. The results remain almost
the same, and the holding company effect is more strongly supported.
We also test a model where we use the changes in family ownership
and affiliated ownership to accommodate the dynamic aspect of the
model, and find that the results remain the same.
