This year is the 20th anniversary of the listings of Sina Corp., Netease and Sohu.com, which were the first companies to complete offshore listings using a structure that would later become known as the variable interest entity structure (the “VIE Structure”). On the occasion of this significant milestone, it seems appropriate to review the history, current status and future prospects of this structure.
For the last two decades, the VIE Structure has been used extensively to facilitate offshore financing of Chinese businesses in the technology, media, and telecommunications (“TMT”) and other regulated industries. A VIE Structure typically involves contractual arrangements pursuant to which an offshore holding company (the “Offshore SPV”) (usually through a wholly foreign-owned enterprise (“WFOE”) established in China) controls and receives the economic benefits of a Chinese onshore operating entity (the “VIE”) whose shareholders would normally be PRC nationals (the “Nominees”).
The VIE holds the assets and licenses that cannot be legally owned by foreign investors or foreign-invested entities. The VIE, the Nominees, and the WFOE enter into a suite of contracts that enable the VIE’s operating results to be consolidated for financial accounting purposes into the financial statements of the Offshore SPV. The VIE Structure represents an attempt to grant to the Offshore SPV and its subsidiaries the rights and benefits normally associated with ownership of the VIE without holding actual equity ownership, thereby enabling foreign investors to invest in regulated sectors in China despite foreign equity ownership restrictions or prohibitions.
A typical VIE Structure chart is set forth below. The Offshore SPV seeks to control the VIE through the contracts entered into by the WFOE with the Nominees (the “Control Contracts”) and obtains the profits of the VIE through the contracts entered into by the WFOE with the VIE (the “Economic Contracts”).
II. THE UNICOM OR CCF MODEL: THE PREDECESSOR TO THE VIE STRUCTURE
What later came to be known as the VIE Structure came to the attention of the world in the year 2000 with the listing in quick succession of Sina Corporation, Netease, and Sohu, each of which utilized an early version of what would later come to be known as the VIE Structure to permit foreign indirect investment in the Internet content provider business.
However, the history of the VIE Structure in China did not start in 2000. In fact, a predecessor to the VIE Structure, commonly known as the Unicom or China-China-Foreign Model (“CCF Model”) was being used as early as the mid-1990s. Unicom is a Chinese telecommunications company that was established in 1994 to be a competitor of China Telecom. When it was established, Unicom had one major problem: it had a very limited existing network and insufficient funds to construct a network that could in any way compete with China Telecom. Unicom soon discovered, however, that foreign telecommunications companies were ready and willing to go to extraordinary lengths to make even a risky and indirect investment in the Chinese telecom market.
In the mid-1990s, China had not yet become a member of the World Trade Organization (“WTO”) (and would not do so until late 2001) but it was already widely anticipated that upon accession, the telecom market would, to some extent, be opened to foreign investment. Foreign telecom companies were anxious to establish a toehold in China in order to be poised for investment once China acceded to the WTO.
The CCF Model, which was adopted by several telecom companies including France Telecom, GTE (the predecessor to Verizon), Deutsche Telecom, and Bell Canada, involved the foreign telecom company establishing a joint venture (“JV”) in China with a subsidiary of Unicom. The JV would then enter into various contractual arrangements with Unicom under which the JV (using funds provided by the foreign telecom company) would construct a network (usually a paging or mobile network) and purchase equipment, all of which would be owned by Unicom.
This structure raised a number of potential issues under Chinese law. Interestingly, however, the one issue that was singled out by the State Council in an internal report commissioned on the CCF Model in the late 1990s related to accounting irregularities. Many of the JVs using the CCF Model were accounting for the depreciation of the network and equipment in their financial statements despite the fact that they did not actually own the network and equipment. As the CCF Model was primarily being used in the asset-heavy paging/mobile network operation sector, this became a material practical issue.
But the demise of the CCF Model was not the result of it being determined to be illegal but rather a result of Unicom needing to jettison its JV partners in order to prepare itself for the Hong Kong listing of its offshore affiliate, which was completed in June 2000. In 1999, China’s telecommunications regulator (then known as the Ministry of Information Industry or “MII” and today known as the Ministry of Industry and Information Technology or “MIIT”) issued a formal notice declaring the JVs to be irregular and requiring their cleanup. Interestingly, the notice did not state that the CCF Model itself was illegal but merely stated that it needed to be rectified. This led to much speculation among market participants that MII had issued the notice at the behest of Unicom.
On the basis of this notice, Unicom approached its JV partners and told them that the JVs needed to be terminated. Each JV partner negotiated with Unicom about how it would be compensated. It is instructive to note, particularly given the current discussions of force majeure provisions in light of the COVID-19 outbreak, that the leverage that each JV partner had largely depended on the exact wording of the force majeure provision in the JV contract that it had entered into with the relevant subsidiary of Unicom. The foreign telecom companies that fared best and got the richest payoffs (including warrants to subscribe for shares in Unicom once listed) were those that had entered into force majeure provisions that did not mention governmental action or excluded governmental action by the agency that supervised the Chinese partner.
This “cleanup” of the Unicom entities left a significant cloud over the future use of CCF Model.
III. THE SINA OR NETEASE MODEL
It might seem odd that Sina, Netease, and Sohu used a structure that arguably descended from the CCF Model to list successfully on foreign stock exchanges in 2000 when the CCF Model had been declared “irregular” by the MII only a year before. However, the model used by Sina, Netease, and Sohu (at first commonly referred to as the Sina or Netease model (the “Netease Model”)) differed from the CCF Model in several key respects.
First, and perhaps most importantly, unlike the CCF Model which had primarily been adopted to build and operate asset-heavy basic telecom services (like mobile and pager networks), the Netease Model was adopted, at least initially, by businesses operating Internet-based content services such as web portals, webmail, and games. Such Internet Content Providers (“ICPs”) delivered value-added services over basic Internet infrastructure operated by other companies. As such, there was little or no need for ICPs to construct networks or to otherwise acquire significant capital-heavy carrier-grade telecoms equipment and therefore no need for the foreign investor to purchase such assets. Thus, the accounting irregularity identified by the State Council as being the problem with the CCF Model did not arise in the case of the Netease Model.
Second, unlike the CCF Model, which only included Economic Contracts and did not include Control Contracts with the shareholders of the licensed operator (i.e., Unicom) through which the foreign investor might control the operation of the licensed operator, the Netease Model included both Economic Contracts and Control Contracts. The inclusion of the Control Contracts in the Netease Model enabled the Offshore SPV to pursue an offshore listing as it was able to consolidate the financial results of the VIE because the VIE was indirectly controlled by the Offshore SPV.
While this control element might be of concern to regulators, at the same time it made the Netease Model very attractive to foreign financial investors who were seeking a means of investing in China’s Internet industry but also needed a path to an exit such as a stock exchange listing. Once the Offshore SPVs in VIE Structures were listed and, in some cases, expanded considerably in size and value, it became difficult for MII and then later MIIT to push regulations that would declare the VIE Structure to be illegal as it would endanger these major influential listed companies, most of which were still controlled by their Chinese founders. The entities comprising the CCF Model, on the other hand, were never intended to be listed but were instead conceived by Unicom as a mere short-term means of cheap financing to be jettisoned once listing was possible.
Third, whereas the CCF Model was used in relation to a basic telecommunications business operated by a state-owned enterprise, the Netease Model was, at least initially, only used in the private sector in connection with value-added telecommunications (“VAT”) business, which most people assumed would, in any case, be opened up to foreign investment with China’s accession to the WTO. In fact, there is evidence that the Netease Model was initially thought of as a temporary structure that would be unwound subsequent to WTO accession when foreign investment in this sector was permitted and the Offshore SPV or its WFOE could acquire the VIE outright. Unlike the “cleaning up” of the CCF Model, this unwinding would not involve kicking out the foreign investors but instead changing indirect investment by the Offshore SPV in the VIE into direct investment such that the foreign investors would become equity investors in the licensed ICP itself.
IV. THE VIE AS A U.S. ACCOUNTING CONCEPT
The term “VIE Structure” actually only came to be used a number of years after the initial listings of Sina, Netease and Sohu. The term “variable interest entity” or VIE was coined by the Financial Accounting Standards Board in Interpretation No. 46 (Consolidation of Variable Interest Entities) (“FIN 46”) issued in January 2003 in response to the Enron scandal. Prior to FIN 46, a parent company was required to consolidate its subsidiary into its financial statements if it had a “controlling financial interest,” which essentially was concerned only with the ownership of voting interests. Enron had kept certain liabilities off its balance sheet by setting up special purpose entities (“SPEs”) in which Enron held no or only a small equity interest and using these SPEs to borrow loans and purchase assets. Public shareholders had no way of ascertaining the true financial status of Enron because these liabilities did not appear on its balance sheet. The extent of these liabilities only became apparent after Enron’s collapse in 2001.
FIN 46 sought to prevent companies from hiding liabilities from shareholders by requiring them to consolidate SPEs used by companies for specified purposes into their financial statements. Under FIN 46, the ownership of voting interests is no longer the sole determinant of “controlling financial interests” where SPEs are concerned, instead, SPEs will be consolidated if the “parent” is exposed to the majority risk of the SPEs’ returns and losses. These SPEs were then called VIEs under the new rules if they met the requirements for consolidation.
This development of accounting standards strengthened the position of companies using the Netease Model. Following FIN 46, they were not simply permitted to consolidate the financial results of companies that they contractually controlled, but were now required to do so under the new accounting rules.
V. WEAKNESSES OF THE VIE STRUCTURE
Over the past two decades, certain weaknesses of the VIE Structure have become apparent. We briefly outline a few of these weaknesses below.
a. Ownership of Mark Under Which Service Is Provided and URL for Telecom/Internet Businesses
In July 2006, MII issued the Circular on Strengthening the Administration of Foreign Investment in and Operation of Value-added Telecommunications Businesses (the “MII Circular”), which, among other things, mandates certain key assets (such as trademarks, domain names, and telecommunication facilities such as routers and servers) required for conducting the VAT business be held directly by the VIEs and the Nominees. Although not expressly referred to in the MII Circular, the VIE Structure is generally believed to have been the target of the MII. Prior to the MII Circular, it was common to see VIE arrangements in which the VIE was essentially a shell. In those arrangements, the WFOE would own essentially all of the technical assets and branding used in the business and make them available to the VIE through services and intellectual property licensing agreements in return for service fees and royalties equivalent to the profits generated by the VIE. As a result of the MII Circular, it became necessary for the VIE to possess more substance. While it is still possible for the WFOE to provide equipment to the VIE by means of a formal operating lease, the VIE is required to employ various type of personnel directly and either the VIE or the Nominees must own the main branding (i.e., the trademark) used by the business. This has had two key consequences. Firstly, by insisting that more of the business assets be held by the VIE, the MII Circular increased the downside risk to shareholders of the Offshore SPV in the event that problems arise with the VIE.
In addition, the requirement that the branding be owned by the VIE or its Nominees made it more difficult for actual foreign Internet companies to operate businesses in China under their own brands through a VIE Structure. Whereas for a Chinese founder controlled Offshore SPV such as Baidu it is a mere inconvenience to have to transfer rights to its brand to its VIE, a foreign Internet company like Google is simply not going to transfer ownership of its key house marks (even in Chinese) to a Chinese entity held by Nominees. For these reasons, many market participants saw the MII Circular as an attempt to limit actual foreign Internet/telecom companies from using the VIE Structure to operate businesses in China.
b. Misalignment Issues
Another practical risk associated with the VIE Structure stems from the potential misalignment of interests between the Nominees and the shareholders of the Offshore SPV. When the Nominees do not have a significant equity interest in both the VIE and the Offshore SPV, or their interests are threatened, or somehow they decide that taking back control of the VIEs is in their best interests, they may be tempted to breach or threaten to breach the Control Contracts and the Economic Contracts.
The story of GigaMedia Limited (“GigaMedia”), a Taiwan-based Nasdaq-listed company, that operated its online game and service business in the PRC through several VIEs, is one such example of this risk. GigaMedia acquired control of these VIEs in 2007 through the acquisition of T2CN Holding Limited (“T2CN”), an Offshore SPV, and left the original founder to continue controlling both the WFOE and the VIEs. In early 2010, GigaMedia decided to remove the founder from his positions at several PRC entities, but the founder refused to step down and took possession of the company seals, financial chops, and business registration certificates of the relevant entities. Without control over the company seals and relevant documentation, GigaMedia was not even able to register the shareholder resolutions to remove the founder from his positions. Although GigaMedia attempted to gain control over the VIEs through court proceedings, it failed and ultimately had to deconsolidate T2CN’s financial results and take extensive write-offs.
As in the case of GigaMedia, the misalignment issue primarily occurs when the VIE Structure is being used by a true foreign company to operate its business in China. Because the foreign company is not controlled by a PRC national, it has no choice but to use mere employees who happen to be PRC nationals (or in the case of GigaMedia, the original founder of acquired entities) as the Nominees to hold the equity of the VIEs. If those employees are fired or otherwise have a falling out with their employer, it is very tempting for them to refuse to transfer the equity of the VIE without being given a significant amount of compensation.
There are ways of structuring the VIE Structure to minimize the risk of misalignment. However, such a risk can never be entirely eliminated.
c. Issues Resulting From Divorce or Death of Nominees
Because the Nominees holding the equity interests in the VIE are generally natural persons, another practical risk associated with the VIE Structure is the potential divorce or death of Nominees.
One well-known example of the divorce of a founder causing issues for a company using the VIE Structure was the delay of the listing of Tudou (China’s equivalent of Youtube) caused by the claim (brought in a divorce court in Shanghai) by the former wife of the founder, chairman, and CEO of Tudou that the portion of the equity of Tudou’s VIE held by the CEO of Tudou as a Nominee constituted community property.
Tudou was effectively forced to delay its IPO for six months until the CEO had reached a settlement with his former wife. This delay, which came at a time when market conditions were deteriorating, very likely resulted in a significant reduction in the valuation at which Tudou was ultimately able to list its shares.
Again, there are ways of structuring the suite of contracts constituting the VIE Structure to mitigate these “natural person” risks, but they can never be completely eliminated.
d. Issues Resulting From the Use of the VIE Structure in Non-Internet Sectors and for Purposes Other Than Avoidance of Foreign Investment Restrictions
Whereas the use of the VIE Structure in the value-added Internet sector has become quite standard in China, its use in other sectors can sometimes raise issues. For example, as indicated above, one key factor differentiating the VIE Structure from its predecessor, the CCF Model, is that the use of the VIE Structure in the value-added Internet sector did not require the foreign investor to funnel significant capital to the VIE as there was no need to acquire significant carrier-grade telecom equipment or to construct a telecom network that would need to be owned by the VIE as the licensed operator.
These issues do arise, however, when the VIE Structure is used with respect to more asset-heavy businesses such as Internet data centres (“IDCs”). There are structures that represent significant variations on the VIE Structure that can be used to invest in IDCs, but a discussion of such structures is beyond the scope of this article.
Use of the VIE Structure in sectors other than telecommunications can raise additional issues depending on the specific legislation in those sectors and the degree to which the PRC government regards these sectors as “sensitive” in relation to foreign investments.
Although for a period of time after the issuance of the Regulations on Mergers and Acquisition of Domestic Enterprises by Foreign Investors (the “M&A Rules”) in September 2006, the VIE Structure was used by some to avoid central level PRC government approval of “round trip” restructuring (that is, the offshoring of purely domestic companies), this use of the VIE Structure was effectively ended in November 2011, when the Hong Kong Stock Exchange (HKSE) updated its listing decision (HKEX-LD43-3) to indicate, in effect, that it would not permit the listing of shares of companies on the HKSE if those companies were using the VIE Structure for any purpose other than the avoidance of restrictions on foreign investments in particular sectors in China.
Thus, unless companies want to rule out from the beginning a listing on the HKSE – which few are willing to do – they can only use the VIE Structure as a means of avoiding restrictions on foreign investments in particular sectors in China and not for any other purpose.
VI. WHY THE VIE STRUCTURE CONTINUES TO BE USED POST-WTO ACCESSION
As noted above, in the year 2000 when Sina, Netease, and Sohu used what would become known as the VIE Structure to list in the United States and Hong Kong, there was an expectation that this structure was only being used temporarily until China acceded to the WTO and foreign investment in the Internet sector was liberalized. However, even though China acceded to the WTO in 2001, foreign investment in ICPs is still limited to no more than 50% of the equity ownership of an ICP JV (after being gradually raised over the course of time since WTO accession) and, perhaps more importantly, in practice it is quite difficult for a JV to obtain such a license and the business scope of a foreign-invested JV with such a license is often required to be much more limited than the business scope that would be granted to a purely domestic ICP license holder.
Until (a) an ICP license holder can be 100% foreign-owned, (b) the permitted business scope of such an ICP license holder is the same as a purely domestic ICP license holder, and (c) the procedure for a foreign-owned entity to apply for an ICP license is no more stringent than that for a purely domestic entity to do so, the VIE Structure is likely to continue to be the most popular route for foreign investment in the Internet sector in China.
In Part II of this article, which will be published in the next edition of Hong Kong Lawyer, we will consider recent developments with respect to the VIE Structure and the legal status and risks related to the use of the VIE Structure as it enters its third decade.
 According to a report issued by the China Academy of Information and Communications Technology in January 2019, only 49 Chinese-foreign JVs have obtained ICP licenses since WTO accession in 2001.