Passive disclosure

In the majority of cases, the NDRC adopted a passive approach in disclosing its investigations. For instance, it revealed its investigations after firms’ self-disclosures in two abuse of dominance cases involving InterDigital (InterDigital Case) and Qualcomm (Qualcomm Case), as well as a resale price maintenance case involving a number of luxury-car manufacturers and their dealers (Auto Part Cases). In a cartel case involving 12 Japanese auto companies (Japanese Auto Cartel Case), the NDRC disclosed its investigations prior to the firms’ self-disclosures, but did not specify the names of the individual companies involved. In two resale price maintenance cases, one involving Maotai and Wuliangye (White Liquor Case) and the other involving premium optical manufacturers (Eye Vision Case), the NDRC announced its decisions directly without any prior disclosure of its investigations. In both cases, companies were seen to have actively co-operated with the NDRC's investigations and agreed to rectify their behaviour before the NDRC announced its decisions. Notably, the decision of Maotai only appeared on Guizhou DRC's website briefly and was then quickly removed overnight. NDRC officials noted that because Maotai is a local state-owned champion, local officials faced significant pressures to announce this case.Footnote ¹⁰ In one case, the NDRC delayed announcing the decision until months after the case was completed. On 2nd September 2014, the NDRC announced that it had imposed a fine of approximately 110 million yuan on a number of insurance companies and an insurance association (Zhejiang Insurance Case).Footnote ¹¹ However, the decision had in fact been made more than nine months earlier.

A few insiders indicated that the NDRC had investigated many more cases than those that it disclosed in public.Footnote ¹² For instance, in April 2009, several media outlets reported that five large domestic state-owned airline companies were suspected of fixing the prices of airfares.Footnote ¹³ The NDRC never made any announcement about any related investigation. But acute observers found that this case was briefly mentioned in an obscure Chinese magazine entitled Price Supervision and Check (Jiage jiandu yu jiancha 价格监督与检查).Footnote ¹⁴ The magazine is not widely circulated and is mostly read by government officials responsible for price monitoring and control. According to the reports in the magazine, the NDRC representatives investigated these domestic airlines and requested rectification of their conduct in August 2009.Footnote ¹⁵ It further disclosed that the NDRC submitted a report of its investigation to the State Council, which was subsequently reviewed by the then vice-premier Li Keqiang 李克强.Footnote ¹⁶ According to an NDRC official, no public sanction was imposed on these airlines because the case was “harmonized,” i.e. resolved inside the bureaucracy.Footnote ¹⁷ Such a phenomenon of “internal dispute resolution” is not uncommon. A scholar's close examination of the reports published in Price Supervision and Check between 2009 and 2013 revealed more mysterious omission by the NDRC of its antitrust investigations.Footnote ¹⁸ These cases involved a wide range of domestic targets, including regional monopolies and large central SOEs. As later acknowledged by Xu Kunlin during a TV interview, the NDRC had kept a low profile of many cases that it had investigated out of concern of “double penalty.”Footnote ¹⁹ He explained that many firms requested the NDRC not to publicize its case for fear of the reputational sanction on these firms.Footnote ²⁰

Proactive disclosure

In contrast to the low-key approach that it took in most cases, on three occasions the NDRC acted as a whistleblower by strategically leaking information to state media. When the NDRC publicizes the case through highly influential state-media outlets such as CCTV and the People's Daily, the media coverage exposes the firm to a high level of publicity.

On 9 November 2011, Li Qing 李青, a deputy director general at the antitrust unit of the NDRC, announced on CCTV that her agency had been investigating two large telecommunication firms – China Telecom and China Unicom – for antitrust violations.Footnote ²¹ Ms Li claimed that the two SOEs had a dominant market position and together held over two-thirds of the market share in the internet access market.Footnote ²² According to her, these two SOEs had conducted price discrimination against rival companies – if the facts were ultimately proven true, these two firms could be subject to a fine of 1–10 per cent of the fiscal revenue, potentially up to 8 billion yuan for both SOEs.Footnote ²³ In addition to the antitrust allegations, Ms Li asserted that the two SOEs had not achieved full integration, thus increasing the cost of internet access for internet service providers.Footnote ²⁴ The CCTV programme also referred to a recent report prepared by the National Telecommunication Expert Committee, noting that China's internet speed ranked 71 in the world, less than one-tenth of the average speed of OECD countries while costing two or three times more.Footnote ²⁵ The CCTV programme showed quoted estimates from the report, noting that the internet access price would decrease by 27 per cent to 38 per cent in five years if the relevant market became truly competitive, saving 10–15 billion yuan for Chinese consumers.Footnote ²⁶ The NDRC announcement caused a sensation in China and public opinion was overwhelmingly negative for the two SOEs. A survey conducted by Sina Weibo, the Chinese version of Twitter, on the day of NDRC's announcement shows that 96 per cent of the participants believed that these two SOEs held dominant positions in the broadband market and that 86 per cent of the participants were dissatisfied with their performance.Footnote ²⁷

As later revealed by Xinhua News, the NDRC's investigation of these two SOEs was met with significant opposition from various ministries who were concerned about the controversy over the matter and urged the NDRC not to reach its conclusion hastily before gathering solid evidence.Footnote ²⁸ Thus, the NDRC's announcement on CCTV not only sent a shock to the SOEs, but also to the other government departments who were involved and consulted before the investigation.Footnote ²⁹ According to Xinhua News, the surprise announcement on CCTV was a deliberate tactic adopted by the NDRC in response to the arrogant attitude displayed by these two SOEs.Footnote ³⁰ Interviews with NDRC officials corroborate the news report. Insiders note that the NDRC faced significant opposition from bureaucratic departments including the State-Owned Assets Supervision and Administration Commission (SASAC), which was concerned about the potential loss of state assets, and the Ministry of Industry and Information Technology (MIIT), which was concerned that NDRC was encroaching on its turf.Footnote ³¹ By leaking the news to CCTV, the NDRC hoped that public opinion against the two SOEs would in turn exert pressure on these two firms, making it more difficult for other bureaucratic departments to intervene in the matter.Footnote ³²

Two years later, the NDRC applied the same tactic again when it investigated a number of domestic and foreign infant-formula manufacturers (Infant Formula Case). On 1 July 2013, CCTV claimed that it had obtained confirmation from the NDRC that the agency had been investigating Biostime and a few other infant-formula manufacturers for conducting resale price maintenance (RPM) in China.Footnote ³³ Notably, CCTV only named Biostime but omitted the names of the other manufacturers. CCTV News specifically highlighted that Biostime had a large market share in the premium infant-formula market and had been charging abnormally high prices for its products. The day after CCTV's announcement, the People's Daily disclosed the names of five more companies which were also involved, claiming that it had obtained confirmation from the NDRC.Footnote ³⁴ Mysteriously, Meiji, Fonterra, and Beingmate, who were at that time also being investigated by the NDRC, were not mentioned in the news. It therefore appears that within the same case, the NDRC exposed the firms under investigation to three different levels of publicity, as shown in Table 2.

Table 2: Disclosure Pattern of Investigation in Infant Formula Case

Source:

Based on official penalty decisions by the NDRC and various news articles about these cases kept on file by the author.

Notes:

*CCTV: indicates whether the firm was mentioned on the CCTV programme.

**People's Daily: indicates whether the firm was mentioned in the People's Daily.

***Severity of penalty: measured by the percentage of the revenue used to calculate the fines imposed on the individual firms.

According to an NDRC official, such differential disclosures have to do with the firms’ level of cooperation rather than the severity of their conduct.Footnote ³⁵ That is, if a firm vigorously defended itself and did not quickly admit its guilt, the NDRC would expose the firm to more negative publicity and impose a higher penalty on the company. In this case, Biostime was the most difficult to deal with, as the firm fiercely defended its conduct during the investigation.Footnote ³⁶ This contrasted with other firms involved in the investigation – an example that the official gave is Meiji, which readily cooperated with the agency and even volunteered to submit all the evidence of its resale price maintenance practices to the agency.Footnote ³⁷

The third case involved a number of gold retailers in Shanghai who allegedly colluded to fix prices for gold retailing (Gold Retailer Case) in 2014. The People's Daily, a Party mouthpiece, was the first to disclose the investigation, claiming that it had obtained confirmation from the NDRC as exclusive coverage.Footnote ³⁸ Similar to the above two cases, the NDRC appeared to have been met with strong resistance during its investigation. During a TV interview, an NDRC official recalled a direct confrontation with the chairman of the trade association, who had orchestrated the cartel.Footnote ³⁹ Notably, some of the gold retailers are partly state-owned and, in fact, the Shanghai government used to own all of them.Footnote ⁴⁰ A lawyer involved in this case noted that some of the gold retailers were initially contemplating filing an administrative appeal against the NDRC but ultimately relented due to the pressures from the government.Footnote ⁴¹

A potential alternative explanation of the NDRC's high-profile announcement in these cases is that the agency was seeking to raise awareness among the Chinese public about antitrust compliance. After all, the AML had only been promulgated a few years ago and businesses and citizens were only starting to learn about the law. A closer look at these cases, however, suggest that education was unlikely to be the main motivation for the NDRC's proactive disclosure. First, the NDRC could have achieved the purpose of increasing public awareness through the publication of its final decisions, which might have achieved a greater deterrent effect, as the case outcome would have been known and the penalty would have been imposed at that point. Moreover, if the primary purpose of such proactive announcement was to educate the public, there would have been no need for the agency to expose the firms subject to its investigation to different levels of publicity, as shown in the Infant Formula Case. Furthermore, interview findings with the NDRC officials in these cases show that they tried to leverage public opinion to push forward difficult investigations. This increased the cost to firms from defying the agency and significantly reduced the NDRC's cost of prosecution.

Thus far, we have only observed three cases in which the NDRC applied the strategy of proactive disclosure. It is possible that in subsequent investigations, companies became aware of the strategic shaming sanction and thus were deterred from challenging the NDRC. Another possibility may have to do with the personal style of the political leader. Xu Kunlin, who is known to use media tactics to push forward difficult cases, left the antitrust bureau in early 2015. The new Director General, Zhang Handong 张汉东, is near retirement and is not as keen to apply the shaming sanction to compel enforcement.Footnote ⁴²

The market model

According to the efficient market hypothesis theory, the price of stocks reflects the value of all publicly available information about the company to investors. Therefore, any news about the fundamentals of the company is immediately absorbed and reflected in the share price. To examine the market reaction to the government's announcement of an investigation, it is important to estimate the stock price that would have prevailed in the absence of the news at the time that the investigation was announced by the NDRC. The difference between the counterfactual return and the actual return represents the abnormal return. Following existing literature on the study of the impact of an antitrust investigation on market value, I use a simple market model of returns to calculate a counterfactual return.

The market model is defined as:

$$R_{it} = \alpha _i + \beta _iR_{mt} + \varepsilon _{it},$$

Where R _it is the return of security i at period t, i.e.

$$R_{it} = \displaystyle{{P_{i,t}} \over {P_{i,t-1}}}-1$$

and R _mt is the return of the market index at period t. The error term ε_it is expected to be mean zero, and the variance equals to σ ²_ε. P _i,t and P _i,t−1 are the closing prices for security i at time t and time t−1 respectively.

The abnormal return can therefore be defined as:

$$AR_{it} = R_{it}-E\lpar {R_{it}{\rm \vert} X_i} \rpar $$

The daily abnormal returns are summarized over the event window to derive the cumulative abnormal returns (CARs):

$$CAR_{{\rm i(}{\rm T}_{\rm 1}{\rm -}{\rm T}_{\rm 2}{\rm )}} = \sum\nolimits_{T_1}^{T_2} {AR_{it}} $$

where CAR_i(T1−T2) is the cumulative abnormal return for firm i over the event window (T2, T1). I examine four event windows: a three-day event window [−1, 1], a seven-day event window [−1, 5], an 11-day event window [−5, +5] and a 31-day event window [−20, +10]. The estimation window dates from T0 (t = −140) to T1 (t = −21) relative to the announcement date t = 0, including 120 trading days.

Stock reactions

The previous literature suggests that the negative abnormal returns experienced by a firm upon the disclosure of an antitrust investigation reflects investors’ prediction of the firm's value loss from three sources: the estimate of the fines that the company would receive, the loss of profits resulting from the price reduction, or other behavioural remedies offered to the regulator, as well as reputation sanctions.Footnote ⁴⁵ In the Chinese context, however, a high-profile public announcement by an antitrust agency sends another important signal to the market – the firm's loss of political capital. In China, political connection with the government is a particularly important and valuable asset for a firm.Footnote ⁴⁶ This is especially the case for SOEs, which have close ties with the Chinese government and enjoy convenient channels for lobbying its various bureaucratic departments.

As set out in Table 3, in both the China Telecom/China Unicom Case and the Gold Retailer Case, companies experienced significant negative abnormal returns either on the day of or the day after the NDRC's announcement. Notably, China Telecom and China Unicom are two of the three largest telecom companies in China and they are both directly owned by the central government. Yu Yuan Tourist Mart and Lao Feng Xiang were both formerly owned by the Shanghai government but the former was privatized, while the latter remained owned by the state. Thus, in addition to investors’ prediction of the loss of fundamental value of these firms, investors may also predict a loss of political capital of these firms. The NDRC's announcement on CCTV or in the People's Daily may signal that the SOEs involved in these cases have fallen out of political favour, hence they lack the political clout to avoid such investigation or to stop the disclosure by the agency in the first place. Indeed, as discussed in the previous section, NDRC officials acknowledged that they did not disclose some cases due to the political pressures that they faced in investigating these firms.

Table 3: Impact on Stock Prices in China Telecom/China Unicom Case and Gold Retailer Case

*** significant at 1%; ** significant at 5%; * significant at 10%.

Unlike the above two cases involving domestic SOEs, firms investigated in the Infant Formula Case were exposed to three different levels of publicity, as set out in Table 1. This allows us to isolate the effects of the disclosure and presents us with a unique opportunity to examine and compare the market reactions in these different scenarios.

In the Infant Formula Case, Biostime was subject to the most publicity and it clearly felt the strongest impact from the NDRC's announcement. Although the company had made a prior disclosure that it had been subject to antitrust investigation, the NDRC's subsequent announcement still sent a very bad signal to the market. Biostime's stock suffered a cumulative negative abnormal return of 22 per cent upon the NDRC's announcement in the three-day event window and 37 per cent in the seven-day event window. Both results are highly statistically significant. This shows that investors can gain new information from the NDRC's action, even though the government's announcement itself contains no more information than Biostime's own disclosure. Because Biostime was the only firm that was named on CCTV, investors may have read this message as a sign that the NDRC had obtained solid evidence of the firm's antitrust violations and that the firm's infringement was very severe. Therefore, the loss of equity may reflect investors’ predictions of the legal sanctions, including the potentially high fine that could be imposed under the AML, as well as behavioural remedies (such as price reductions) formally or informally imposed by the NDRC. Furthermore, the NDRC announced the news on a CCTV programme which condemned the excessively high prices of Biostime's products, and in subsequent days Biostime found itself lambasted by a slew of Chinese media accusing it of using aggressive distribution tactics to reap abnormal profits. Investors may therefore predict a loss from the reputation sanction – that the overwhelming public criticisms targeted at Biostime would result in a loss of trust from consumers and suppliers, thus affecting the firm's sales and future performance.

In addition to Biostime, Mead Johnson, which was mentioned in the People's Daily, also suffered a huge loss upon the announcement. Its stock suffered a cumulative negative abnormal return of 5 per cent upon the NDRC announcement in the three-day event window and 13 per cent in the seven-day event window. Both results are statistically significant. However, the stock reactions of Abbott, the holding companies of Dumex (i.e. Danone) and Wyeth (i.e. Nestlé), which were also mentioned in the same newspaper, are not as strong. This may have to do with the fact that these companies are large multinational conglomerates and the Chinese investigation only affected a small part of their global businesses. Based on the disclosure from annual reports, Abbott's revenue from China accounted for 5 per cent of its global revenue in 2013, and the Chinese revenue of Danone and Nestlé only accounted for 7 per cent of their overall revenue during that year. Moreover, milk powder is not a major source of revenue for these firms. For instance, income from whole-milk powder business accounted for only 10 per cent of the total revenue of Abbott, 20 per cent of Danone, and 14 per cent of Nestlé in 2013. This contrasted with Mead Johnson, which invested heavily in China and derived almost 31 per cent of its global revenue from the mainland market in 2013.Footnote ⁴⁷ Lastly, and not surprisingly, for those firms that were omitted in the news (i.e. Fonterra, Beingmate and Meiji), the NDRC's announcement had little negative effect on stock prices. Notably, after the NDRC's initial announcement on CCTV, there was much speculation about which infant formula producer was involved. As the focus of the inquiry of this article is on the impact of the NDRC's strategic shaming, the market reactions to other types of public disclosures are less relevant here.

To further investigate the impact of differential disclosure on stock prices, I compared the stock reactions of Biostime and Beingmate, the two domestic firms that were subject to two extreme levels of publicity (one was mentioned in both the CCTV programme and the People's Daily, while the other was completely omitted in the news). The business operations of these two firms are highly similar. Both companies sold exclusively to the mainland Chinese market and both derived the vast majority of their revenue from the sales of infant formula. In particular, milk powder accounted for 93 per cent of Beingmate's revenue in 2013Footnote ⁴⁸ and 82 per cent of the revenue of Biostime in 2013.Footnote ⁴⁹ The former is listed on the Hong Kong Stock Exchange and the latter is listed on the Shenzhen Stock Exchange. Thus, it seems that the NDRC's investigation would affect a significant portion of the businesses of both firms. Yet despite their similar exposure to the NDRC's investigation, their stocks had completely different reactions, as demonstrated in Table 5.

On the second trading day after the NDRC's announcement (T = 1), Biostime experienced an abnormal return of −12 per cent and a further return of −19 per cent on the third day. Both results are highly statistically significant. Beingmate, which was also investigated but omitted in the NDRC's disclosure, had a completely different fortune. Its stock prices reacted positively to the NDRC's announcement: the abnormal return on the first day of the announcement was 7 per cent, followed by 5 per cent and 10 per cent on the second and third trading days. All these results are statistically significant. This suggests that the market was probably unaware of Beingmate's involvement at that time. Because Beingmate is a domestic company, the market might have been mistaken in thinking that the investigation was solely targeted at foreign firms and thus domestic firms could have benefited from the NDRC's intervention. Beingmate's stock prices only started to tumble on 4 July 2013, when the firm publicly admitted its involvement in the case.Footnote ⁵⁰ Indeed, Beingmate's delay in disclosing its involvement prompted some Chinese investors to accuse the firm of defrauding investors in violation of the disclosure requirements under Chinese securities law.Footnote ⁵¹

On 7 August 2013, the NDRC released its decision and imposed the heaviest sanction on Biostime. Despite this, the market reacted positively to the news and the firm experienced a positive abnormal return of 7 per cent on the day of the decision. This seems to suggest that the market had earlier overreacted to the NDRC's announcement and had probably over-estimated Biostime's legal sanctions. As one executive involved in the Infant Formula Case told the newspaper after the NDRC's announcement of its investigation: “We hope the NDRC will release its decision soon. The sooner it decides, the less impact it will have on our stock performance.”Footnote ⁵² At the same time, Beingmate experienced a slight positive abnormal return of 1 per cent on the day of the NDRC's decision, followed by a return of −4 per cent the next day, even though the company received full leniency and was not subject to any fine. This is another sign that the market was probably unaware of Beingmate's involvement in the investigation and the negative stock reaction reflects the investors’ expectation of the loss of profits as a consequence of the need to reduce prices after the investigation.

To be sure, the NDRC officials sometimes disclosed their investigations during their press conferences or interviews. But such types of disclosures occurred after the self-disclosures by the investigated firms, and they were not publicized in highly influential media outlets such as CCTV or the People's Daily. Thus, the NDRC did not intend to employ the strategic shaming sanction in those circumstances of passive disclosure. For completeness, I investigated the stock-market reactions of those cases in which the NDRC took a reactive approach in disclosing its investigation. As shown in Table 6, the market reactions to the NDRC's announcements in these cases are much weaker than those observed in Tables 3 and 4. Because the NDRC did not specify the firm involved or made no disclosure through other types of passive disclosure, I cannot conduct any meaningful event study in those circumstances.

Table 4: Impact on Stock Prices in the Infant Formula Case

*** significant at 1%; ** significant at 5%; * significant at 10%.

Notes:

1. The NDRC announced its investigation of the Infant Formula Case on 1 July 2013, a public holiday in Hong Kong. Therefore, T = 0 for Biostime is 2 July 2013.

2. The companies listed in Table 4 are listed on different stock exchanges in Hong Kong, Paris, Shenzhen, New York and Tokyo. For the purpose of examining their abnormal return, I use the stock price of their holding companies. Thus, for the purpose of analysing the abnormal returns of Dumex and Wyeth, I use the stock returns of their holding companies, Danone and Nestlé, respectively.

Table 5: Comparison of Abnormal Return and Cumulative Abnormal Return of Biostime and Beingmate

*** significant at 1%; ** significant at 5%; * significant at 10%.

Note:

Because 1 July 2013 is a public holiday in Hong Kong, T = 0 is 2 July 2013 for Biostime.

Table 6: Impact on Stock Prices upon the NDRC's Reactive Public Announcement of Investigation by Company and Case Name

* significant at 10%.

Note:

The companies listed in Table 6 are listed on different stock exchanges in New York, Shanghai and Frankfurt. For the purpose of examining their abnormal return, I use the stock price of their holding companies. Notably, FAW-Volkswagen is a joint venture between FAW Group and Volkswagen so their stock returns are presented separately here.

To further assess the impact of the announcement on a firm's value, Table 7 presents the loss of market capitalization during a three-day event window upon the NDRC's announcement. It is difficult to evaluate precisely the reputation effects in these cases, but it is clear that the ultimate fine imposed on these firms is dwarfed by the value loss that these firms experienced upon the regulatory announcement. To illustrate, Biostime suffered an equity loss of almost 4.39 billion yuan upon the NDRC's announcement during a three-day event window, which is 26 times larger than the fines that it ultimately received.

Table 7: Comparison of Value Loss with Fine during Three-Day Event Window upon the NDRC's Announcement

*** significant at 1%; ** significant at 5%; * significant at 10%.

Notes:

1. Market capitalization is calculated as the product of share price and the number of outstanding shares.

2. The value loss is calculated by multiplying the cumulative abnormal returns over the [−1, +1] event window by the market capitalization of the firms on the trading day before the event.

3. The value losses of Biostime and Mead Johnson have been converted into yuan using the historical exchange rate on the event date.

4. The China Telecom/China Unicom Case was excluded as the case was ultimately suspended and the NDRC did not impose any fine.

As the NDRC possesses significant discretion in deciding whether, when, and how to disclose an investigation, it can potentially use this discretion to its advantage. If a publicly listed firm does not cede to the agency's demand, it could face a strategic proactive disclosure by the antitrust agency, resulting in a dramatic equity loss. On the other hand, if a firm readily gives in, the agency might not disclose its investigation, might delay announcing its decision, or might even conceal its decision, thus minimizing the negative impact on its stock performance. Such a strategic move creates powerful incentives for firms to cooperate with the agency and deters them from defending themselves or challenging the agency in public or in court.

Importantly, these early precedents in Chinese antitrust enforcement have become the focal pointFootnote ⁵³ for future cases. Lawyers and in-house counsel used them as lessons to “educate” clients as to how they should cope with an NDRC investigation and to warn them against the consequences of a vigorous defence.Footnote ⁵⁴ In subsequent investigations, particularly in those resale price maintenance cases, manufacturers proactively offered significant price reductions in response to the NDRC's investigations in order to show willingness to cooperate with the agency.Footnote ⁵⁵ While this gesture would have been deemed highly risky in other jurisdictions and, indeed, has puzzled foreign antitrust experts,Footnote ⁵⁶ it seems to be a rational decision in the Chinese context. Given the significant discretion that the NDRC possesses, a business subject to investigation fears not only a potentially higher legal sanction as a consequence of voicing defence, but also a significant market loss if the NDRC imposes a public shaming sanction. As such, few foreign multinational firms subject to the NDRC's investigation dared to challenge its decisions in court, despite the many grievances that they have voiced through their chambers of commerce.Footnote ⁵⁷