Family Conglomerates in Asia Explained

A family conglomerate, in the Asian sense, is a group of legally separate companies in many industries that are tied together by the same controlling family, usually through a web of shareholdings rather than one merged corporation. The Korean chaebol is the best-known example. But the pattern shows up across the region: Japan's pre-war zaibatsu, India's business houses, the diversified groups of Southeast Asia, and the family-run groups of Hong Kong and Taiwan.

This article is a spoke of the Asia section of the family business library. It explains what these groups are, how they're structured, why economists think they exist, and why the research is split on whether they help or hurt the economies around them. For the wider picture, start with the hub on the role of family business in Asian economies.

Key Facts

• Khanna and Palepu argued in 1997 that the large, diversified business group is the dominant enterprise form across most emerging markets (Harvard Business Review, 1997).
• A 2007 survey in the Journal of Economic Literature concluded that business groups respond to differing economic conditions and can act as either "paragons" or "parasites" from a welfare standpoint (Khanna and Yafeh, 2007).
• A study of Indian business groups found a significant amount of tunneling, much of it through nonoperating components of profit (Bertrand, Mehta and Mullainathan, 2002, QJE).
• When a Korean chaebol-affiliated firm makes an acquisition, its stock price falls on average, while controlling shareholders gain through value created elsewhere in the group (Bae, Kang and Kim, 2002, Journal of Finance).
• Korea prohibited the holding company system in 1986 and reinstated it in 1999; circular shareholding loops among regulated groups fell from 282 at ten groups in May 2017 to 11 at two groups in July 2019 (OECD, Duties and Responsibilities of Boards in Company Groups).
• Japan's zaibatsu have been described as pyramidal business groups that coordinated and financed growth in the Meiji era (Morck and Nakamura, NBER Working Paper 13171).

What an Asian Family Conglomerate Is

Researchers usually call these structures business groups. Khanna and Yafeh define them as diversified groups of legally independent firms operating across diverse industries, and note that they're widespread in emerging markets. They also stress that groups share certain attributes but differ a lot in structure and ownership (Khanna and Yafeh, 2007).

Two features make an Asian family conglomerate different from a Western holding company or a conglomerate like the ones built by US acquirers in the 1960s:

1. Separate legal entities, shared control. Each affiliate has its own board, accounts and often its own minority shareholders. The family holds the group together through ownership stakes, cross-holdings, board seats and sometimes a central staff, not through a single corporate parent that consolidates everything.
2. Family at the center. Founding families usually hold the decisive stakes and appoint the senior managers, often across generations. The group's strategy tends to follow the family's priorities as much as any single company's.

The result is a structure that looks like a single enterprise to its customers, bankers and employees, but like a dozen separate companies to securities regulators.

The Main Group Types by Country

The labels differ by country, and so do the details. Use the table as a map, not as a rulebook. Each system evolved under its own history and law.

The table is a simplified map. The groups with direct research evidence are covered below. For Southeast Asia in more depth, see family business in Southeast Asia.

Chaebol: Korea's family groups

Chaebol are Korea's large family-controlled groups. The evidence base is strongest here because their listed affiliates are public and well studied. Bae, Kang and Kim looked at mergers by chaebol-affiliated firms and found that stock prices of acquirers fell on average when a chaebol member made an acquisition, while the controlling shareholders ended up better off because the deal added value inside the group (Bae, Kang and Kim, 2002). That pattern is the core of the tunneling debate, which gets its own section below.

Zaibatsu and keiretsu: Japan's two eras

Japan offers a natural experiment in how groups change. Before World War II the zaibatsu were pyramidal business groups. Morck and Nakamura argue that they provided private-sector coordination of growth in the Meiji era after state-led efforts failed, and that this kind of group-led development works only under conditions that include openness, rule of law, separation of state and business, and, in their words, timely dissolution. Without those conditions, they say, growth stalls and oligarchic families become entrenched (Morck and Nakamura, 2007).

The post-war keiretsu are a different animal. They descend from the group idea but without the family-held top of the pyramid. If you're deciding whether a "Japanese conglomerate" is a family business, the answer depends on the era and the group.

India, Southeast Asia, Hong Kong and Taiwan

Indian groups are the subject of some of the best empirical work, covered below. For Hong Kong, the Claessens, Djankov and Lang study of roughly 3,000 listed companies in nine East Asian economies in December 1996 found that the ten largest families controlled 57.7% of market capitalization in Indonesia, 52.5% in the Philippines, 46.2% in Thailand and 32.1% in Hong Kong (World Bank Private Sector note No. 195, 1999). Those figures are from 1996 and describe ownership at that time, not today's rankings.

How the Structures Work

The mechanics matter, because they explain both why families like the structure and why minority shareholders worry about it.

Pyramids. A family owns a controlling stake in company A, which owns a controlling stake in company B, which owns a stake in company C. The family controls C while owning only a fraction of its cash flows. Morck and Nakamura describe the zaibatsu as pyramidal groups, and the East Asian ownership literature uses the same term (Morck and Nakamura, 2007).

Cross-shareholdings and circular shareholding. Affiliates hold each other's shares. In a circular pattern A owns B, B owns C and C owns A, so a small family stake can control a large web of companies. Korea has regulated this directly, as covered in the reform section below.

Holding companies. A holding company sits on top and owns the affiliates. It's a cleaner structure than a web of cross-holdings. The library's article on the family holding company covers the design choices in detail.

The control-ownership wedge. This is the gap between the control rights a family holds and the share of cash flows it owns. Bertrand, Mehta and Mullainathan frame the problem directly: owners can tunnel resources from firms where they have low cash flow rights to firms where they have high cash flow rights (Bertrand, Mehta and Mullainathan, 2002). The wider the wedge, the bigger the temptation. For how ownership gets spread across a family over time, see family ownership structures.

Why These Groups Exist: The Institutional Voids Argument

The most influential explanation comes from Tarun Khanna and Krishna Palepu. Their 1997 article argues that diversified groups are the dominant enterprise form across most emerging markets, and that focused strategies suited to developed economies can be the wrong model where institutions are weaker (Khanna and Palepu, 1997).

The logic, often summarized as groups filling "institutional voids," runs like this. In a mature market, specialized intermediaries do the work of connecting buyers, sellers, capital and talent: credit bureaus, equity analysts, venture investors, headhunters, courts that enforce contracts. In an emerging market, those intermediaries may be thin or absent. A group fills the gap internally. It can move capital from a cash-rich affiliate to a new venture, transfer managers between companies, lend its brand name to an unproven product, and rely on family trust where contracts are hard to enforce.

Seen this way, the family conglomerate is a rational answer to a missing market, not an oddity. That also implies groups should matter less as institutions improve. Khanna and Yafeh pay particular attention to the circumstances under which groups emerge and to historical evidence on that question (Khanna and Yafeh, 2007).

This is also why diversification looks so different here. A Western analyst may see an unfocused conglomerate trading at a discount. The group's own view is that it's substituting for markets that don't work yet. The library's family business diversification article covers the Western-context trade-offs, and corporate strategy covers the general logic of diversified portfolios.

Paragons or Parasites?

The title of Khanna and Yafeh's survey captures the central debate. Their conclusion is that business groups are responses to varying economic conditions, and that from a welfare perspective they can function as either "paragons" or "parasites" (Khanna and Yafeh, 2007).

The case for paragons: groups substitute for missing markets, finance new industries, and in Meiji Japan arguably coordinated growth that the state couldn't (Morck and Nakamura, 2007).

The case for parasites: groups can entrench founding families, block competitors and move value from minority shareholders to controlling ones. Morck and Nakamura themselves attach a condition: groups work as growth engines only when they can be dissolved or displaced in time, otherwise oligarchic families become entrenched.

The honest summary is that the answer depends on context, on the period, and on how well the minority shareholders are protected. That's why the tunneling evidence matters.

The Tunneling Evidence

Tunneling means moving resources out of a company for the benefit of a controlling shareholder, through below-market asset sales, favorable loans, transfer pricing or similar channels. Two studies are the standard references.

India. Bertrand, Mehta and Mullainathan developed a general method to measure tunneling. It rests on testing the distinctive implications of tunneling for how earnings shocks propagate across firms inside a group. Applied to Indian business groups, they found a significant amount of tunneling, much of it through nonoperating components of profit, such as items outside a firm's core operations (Bertrand, Mehta and Mullainathan, 2002).

Korea. Bae, Kang and Kim studied acquisitions by chaebol affiliates. Acquirers' stock prices fell on average, yet the controlling shareholders gained, because the acquisitions added value elsewhere in the group. The authors read this as support for the tunneling hypothesis (Bae, Kang and Kim, 2002).

Two cautions apply. These studies document specific channels and samples, not every group in every year. And tunneling is a risk created by the wedge, so a group with a narrow wedge, strong boards and active minority investors faces less of it. A family can run a group with integrity. The structure just makes the alternative easier.

Reform: What Korea Did

Korea is the clearest case of a government trying to reshape its family groups, and the OECD record gives the sequence. Korea prohibited the holding company system in 1986, then reinstated it in 1999 to improve ownership transparency and management efficiency. The idea was that holding companies would give a simpler structure than a tangle of cross-holdings. As of September 2018, about one-third of the regulated groups, 22 of them, had converted to holding company structures. Proposed 2018 amendments raised ownership thresholds for subsidiaries from 40% to 50% for unlisted firms and from 20% to 30% for listed ones (OECD, Duties and Responsibilities of Boards in Company Groups).

The same source notes that circular shareholding loops among regulated groups fell from 282 at ten business groups in May 2017 to 11 at two groups in July 2019. Regulation targeted groups with assets of KRW 10 trillion or more.

For readers outside Korea, the lesson is about design. Reform works on three levers: simplify the ownership chain, limit circular holdings, and require more transparency. A family that wants to keep control without those tangles can use a holding company and a family-level governance structure, which is the territory of the family holding company article.

What Families and Advisers Should Take From This

• Know the wedge. If you control more than you own, minority investors will price that gap. A transparent structure lowers that cost.
• Treat each affiliate as a company with its own owners. Related-party transactions need arm's-length terms and independent review.
• Plan for institutions improving. The case for the group is strongest where markets are thin. As those markets develop, the group has to earn its place by outperforming the alternatives, not by default.
• Don't lean on relationships alone. Group advantages often rest on trust networks. The article on relationship networks in Asian family business looks at what those are worth and where they fail.

Related Reading

• The role of family business in Asian economies
• Family business in Southeast Asia
• Relationship networks in Asian family business
• The family holding company
• Family business diversification