key: cord-0072603-3jrqqn7o
authors: Xi, Chao
title: Shareholder Voting and COVID-19: The China Experience
date: 2021-11-03
journal: nan
DOI: 10.1093/cjcl/cxab010
sha: 0ff84c2ebf9074aa1abdce35f0026c57edbd3193
doc_id: 72603
cord_uid: 3jrqqn7o

Using state-of-the-art data-mining techniques, this research constructs a unique dataset comprising the votes cast by Chinese shareholders on 15,553 resolutions laid before 2,888 general meetings of Shanghai Stock Exchange-listed companies in the six-year period between 2015 and 2020. This research empirically depicts, for the first time, the manner in which minority shareholders exercise their rights to vote in China, presenting a counter-thesis to the conventional wisdom of shareholder passivity. In particular, this research offers a unique perspective on how the COVID-19 pandemic has influenced Chinese shareholders’ voting behaviour by focusing its empirical investigation on the 76-day time window from 23 January to 7 April 2020, during which period Wuhan, the epicentre of the COVID-19 outbreak in China, came under a mandatory lockdown order. Our findings offer strong empirical evidence that Chinese shareholders cast their votes in a characteristically more informed manner in the 2020 sample period than they did during the sample periods in the previous five years. The research highlights the potential of shareholder activism in economies where share ownership is concentrated. It also has implications for the ongoing discourse on virtual shareholder meetings and the shareholder franchise.

One of the foundational premises of corporate law and governance scholarship is the rational passivity of shareholders. Conventional wisdom holds that it is only economically rational for a minority investor owning a tiny fraction of shares in a Berle-Means firm to be passive. 1 To be sure, the recent rise of institutional investors offers a unique perspective on how the COVID-19 pandemic has influenced Chinese shareholders' voting behaviour, by focusing its empirical investigation on the 76day time window from 23 January to 7 April (sample period). It was during this window in 2020 that Wuhan, the epicentre of the COVID-19 outbreak in China, was locked down on the orders of the Chinese government in an attempt to rein in the rapid spread of the coronavirus. Our findings offer strong empirical evidence that Chinese shareholders cast their votes in a characteristically more informed manner in the 2020 sample period than they did during the sample periods in the previous five years. The study argues that the COVID-19 pandemic presented a unique setting in which the costs of informed voting were lower, and the benefits of doing so higher, than in normal circumstances. The shift in the cost-and-benefits equilibrium of shareholder voting during the pandemic highlights the potential of shareholder activism in economies where share ownership is concentrated. Otherwise apathetic shareholders in these economies can become considerably less passive, if not active, under such facilitative conditions as the widespread use of state-of-the-art remote technologies that lower the costs of monitoring the firms in which they have invested.

This study also makes contributions to two other strands of scholarship. First, it enriches the extensive empirical literature on shareholder voting. Since the early 2000s, a rapidly growing body of studies has empirically examined how votes cast by shareholders bear on corporate law and governance in the USA, 10 the United Kingdom (UK), 11 various European jurisdictions, 12 and other countries. 13 The case of China, however, has been almost ignored in this burgeoning literature. To be sure, Chinese voting data are publicly available, as will be noted later in this article. Nevertheless, no prior attempt to systematically obtain such data seems to have been made, possibly due to the various technical challenges of reliably extracting data from the voluminous company disclosures of Chinese listed firms. Using state-of-the-art data-mining techniques, this study is among the first to extract Chinese shareholder voting data in a systematic manner. Drawing on these data, this article presents previously unknown evidence on shareholder turnout and resolution-level approval in the Chinese listed sector. It empirically shows that China is closer to its European counterparts than to the USA in these respects. The study further illuminates the unique voting patterns of Chinese minority shareholders (including retail investors) as a group, adding to the emerging line of research on retail shareholder voting. Second, this study also directly engages with the ongoing discourse on virtual shareholder meetings and the shareholder franchise. The early months of 2020 witnessed greater (albeit perhaps temporary) receptivity to the notion of holding shareholder meetings virtually by many jurisdictions that had traditionally restricted, if not prohibited, such an option. 15 The COVID-19 pandemic thus provides an ideal setting in which to empirically assess the impact of virtual meetings on the shareholder franchise. Recent empirical studies present illuminating evidence suggesting that virtual shareholder meetings during the spread of the pandemic are associated with lower shareholder turnout 16 and, perhaps more concerning, with more limited communication between companies and shareholders. 17 It is tempting to infer that the observed shareholder disfranchise is attributable to the move towards virtual meetings. However, changes in shareholder behavioural patterns can also be explained by the pandemic itself, 18 a proposition to which this research lends support. Our research shows that COVID-19 alone largely accounted for the changing behavioural patterns of Chinese shareholders. In other words, the pandemic matters. Our findings therefore caution against hastily drawing any conclusion based on the holding of virtual meetings during the pandemic. In particular, there is little support for the view that a move towards virtual shareholder meetings would do a disservice to the shareholder franchise. 19 The remainder of the article is structured as follows. The second part demonstrates the very limited impact of the COVID-19 pandemic on the People's Republic of China (PRC)'s pre-existing legal framework for shareholder general meetings and shareholder voting. The third part depicts the various types of minority shareholders in Chinese domestic listed companies. It sets the stage for the fourth part, which empirically examines the voting behaviour of Chinese shareholders-in particular, minority shareholders, during the sample periods between 2015 and 2020. We find that Chinese minority shareholders voted in a more informed and intelligent manner in the 2020 sample period, when China was grappling with the nation-wide COVID-19 outbreak, than they did in the five previous sample periods. The fifth part proceeds to offer an evidence-based 15 To help put these measures in context, this part of the article offers a synopsis of certain key aspects of PRC law governing shareholder meetings and voting in respect of listed companies. It will be shown that, unlike in many other jurisdictions, China's extant company law regime, insofar as shareholder meetings and voting are concerned, demonstrates considerable COVID-19 'readiness' and has undergone little change.

Under PRC law, there are two types of shareholder meetings for the Chinese listed company: the annual general meeting (AGM) and the extraordinary general meeting (EGM). 24 That said, the notion of using modern technology to enable shareholders to attend a general meeting remotely long ago gained a foothold in China's corporate governance regime. Under the general rubric of 'facilitating shareholders' participation in general meetings', China's securities regulator has, since 2006, encouraged Chinese listed companies to enable shareholders to attend such meetings by either 'Internet-based means that are secure, economical, and convenient' or by other means. 28 This 'encouraging' approach has become mandatory since 2014. 29 All Chinese listed companies must now supplement in-person general meetings with a technology-enabled component that allows for remote participation by shareholders.

The emphasis of China's regulatory requirements in respect of remote attendance has been predominantly on allowing participating shareholders to vote electronically. Chinese listed companies have been required, since 2014, to enable online voting. 30 Rather than giving individual listed companies the freedom to run their own online voting or to contract it out to a private provider, 31 electronic voting has been administered in a centralized manner by the Shanghai and Shenzhen Stock Exchanges. The Shanghai Stock Exchange therefore runs an online voting platform by which all companies listed and traded thereon are required to conduct online voting pertaining to their shareholder meetings. 32 The Shenzhen Stock Exchange also hosts its own online voting platform. 33 Chinese listed company pays a relatively modest fee on a per-meeting basis to the Exchange-run voting platform to use the platform to organize its online voting. 34 Shareholders who vote via the platform are not required to pay any fee at all. 35 To be sure, remote attendance of shareholder meetings under PRC law can take other forms of electronic communications, ranging from the more traditional audio teleconferencing to the more modern forms of video conferencing, live webcasting, live streaming, or any other similar state-of-the-art virtual meeting technology. In fact, the Shenzhen Stock Exchange spearheaded an Exchangeinitiated pilot scheme back in 2013, using live webcasting for general shareholder meetings. Accordingly, a number of companies listed on that Exchange reportedly live webcast their in-person meetings in February 2013. 36 The pilot scheme was reportedly only intended to be optional. 37 It seems that the practice of live webcasting general meetings, and indeed other virtual meetings, has not hitherto gained much traction in the Chinese listed sector. Our comprehensive database of shareholder voting in China records very few virtual meetings, 38 defined as meetings wherein shareholders who attend remotely via the Internet are allowed not only to vote but also to engage with management, raising questions and voicing their opinions. The few virtual meetings that have been identified appear to have been held in extraordinary circumstances where there was no feasible alternative. For instance, some shareholders attended via video conferencing the shareholder meeting of a Shenzhen Stock Exchange-listed firm while the region in which the company is located was affected by Typhoon Mangkhut, one of the most devastating typhoons on record in that region. 39 It is, nevertheless, fair to say that virtual meetings remain far from common in the Chinese listed sector.

As such, PRC law and practice regarding remote participation in general meetings can be seen as a weak form of virtual participation. The emphasis is almost entirely on enabling shareholders to vote electronically rather than on more elaborate interaction, such as the 'real-time transmission of the general meeting' or 'real-time two-way communication enabling shareholders to address the general meeting from a remote location'. 40 To be sure, shareholders who vote electronically via the Exchange-hosted online voting platform are deemed to have attended and voted at the shareholder meeting. 41 Votes cast 34 Interviews with senior executives of Chinese listed companies. 35 electronically count towards the total votes cast. This approach is in line with the commonly recognized principles on remote participation in shareholder meetings under US law 42 and European Union law. 43 On the face of it, the COVID-19 pandemic presented a formidable challenge to the PRC's legal requirement that general meetings be conducted in-person at a physical location, as the gathering of individuals in a (presumably) indoor environment carries a significantly increased risk of virus transmission. By comparison, many jurisdictions reacted to the pandemic by waiving their legal requirements for in-person general meetings and allowed, at least temporarily, virtual-only meetings 44 as an exceptional measure to limit the risk. These decisions did not always go unchallenged. 45 The PRC authorities did not, however, join this global move to exempt the requirement for in-person meetings. Both the Shanghai Stock Exchange 46 and the Shenzhen Stock Exchange 47 released circulars reiterating their pre-existing position that shareholders should be encouraged and 'guided' to vote electronically in the interest of 'preventing the spread of viruses and protecting the health of investors'. 48 As a gesture of support for that position, both Exchanges waived the fees payable to their online voting platforms for all listed companies incorporated in Hubei Province, the epicentre of the COVID-19 outbreak in China. 49 Interestingly, the Shenzhen Stock Exchange hinted that in-person meetings should be avoided until the pandemic was brought under control, as its circular stipulated that precautionary measures should be taken if there was truly a need to hold an in-person general meeting at a physical location. 50 However, both circulars fell far short of dispensing with the physical meeting requirement. China's resistance to the idea of virtual-only meetings is likely explicable by the fact that, because electronic voting is already mandatory for Chinese listed companies, there is less pressure to move towards fully virtual-only meetings than there might otherwise be.

Under PRC law, a company's AGM must be held annually, 51 normally within the six-month period following the end of its accounting year. 52 required to be held within two months of the occurrence of any of the circumstances prescribed by the PRC Company Law, 53 among other things, where it is deemed necessary by the company's board of directors. 54 A listed company that fails to hold an AGM or EGM within the prescribed period is required to make a report to both the regulatory authorities and the stock exchange on which it is listed, explaining the reasons for the delay. 55 This report must also be made public by way of an announcement by the company concerned. 56 A notice of the meeting must be served 20 days in advance in the case of an AGM, and 15 days in the case of an EGM. 57 Once announced, the AGM or EGM must not be postponed or cancelled without 'due cause'. 58 In the case of postponement or cancellation, the convener of the AGM or EGM-typically, the board of directors-must make an announcement no later than two working days prior to the meeting, providing reasons for the postponement or cancellation. 59 As noted above, the pandemic appears to have disrupted the planning, organization, and conduct of physical, in-person general meetings, as required under PRC law. Comparative studies show that many jurisdictions have granted relief on such formalities as meeting and notice periods to neutralize the impact of COVID-19 on general meetings. 60 By contrast, with the one exception discussed immediately below, the PRC authorities did not generally dispense with preexisting notice requirements or the statutory obligations to hold AGMs and EGMs within the specified time periods. This seeming inflexibility is likely explicable by the fact that the pandemic was universally considered to be the kind of 'due cause' that justifies the postponement or cancellation of a general meeting by its convener.

One major exception to the observed inflexibility is a rule under the aforementioned Joint Circular that relaxes the notice requirement in the specific context of asset restructuring. In the Chinese securities market, asset restructuring has emerged as an important means by which financially struggling listed companies can gain easier access to funding and improve their cash flows during the recent slowdown in China's economic growth. 61 In general, a six-month notice period applies to the EGM that must be held to approve a proposal for asset restructuring after it has been announced by the company undergoing the restructuring. 62 This notice requirement reflects the time sensitivity of a restructuring in normal 53 PRC Company Law (n 24) art 100. 54 Ibid. 55 2016 GM Rules (n 25) art 4(2). 56 circumstances. The pandemic has created considerable uncertainties that can disrupt and delay the process of restructuring. 63 Accordingly, the Joint Circular provides that the six-month notice period can be postponed by one month upon an application by the listed company undergoing the restructuring, and that such an application can be made up to three times, 64 in effect extending the normal six-month notice period by up to nine months. This exception seems to reflect the PRC authorities' determination to ease access to financing through restructurings in an economy that has been further slowed down by the impact of the pandemic. 65

Minority shareholders: who are they in Chinese listed companies?

In the previous part, we offered a synopsis of China's pandemic-related regulatory response to shareholder meetings and voting. This part sets out a broadbrush account of Chinese institutional investors and retail shareholders, respectively, with a view to setting the stage for the discussions in the ensuing parts. Minority shareholders are generally defined under PRC law as shareholders who hold 5 per cent or less of the outstanding shares in a listed company. To put this definition into context, concentrated ownership is a common feature of the Chinese listed sector, as has been demonstrated by a wealth of prior research. 66 Our own analysis shows that, at the end of 2019, the single largest shareholders of the 1,980 Chinese firms listed on the main boards of the two official exchanges owned, on average, 35.93 per cent (median: 33.53 per cent) of outstanding shares. 67 Further, 59.95 per cent of mainboard-listed companies, or 1,175 firms, were under the firm control of a single controlling shareholder that, alone, possessed 30 per cent or more of the company's shares. 68 The identity of minority shareholders can vary greatly from one listed company to another, ranging from substantive shareholders affiliated with the controlling shareholder to small retail investors. Broadly speaking, two groups stand out as particularly relevant to this study: institutional investors and retail 63 77 At the firm level, the coverage of the Ashare market by these funds was remarkably broad. They held stakes in 1,826, or over 97 per cent, of the total 1,980 mainboard-listed companies, with an average of 2.75 per cent (median: 0.80 per cent) of the shares in their portfolio companies, at the end of 2019, as our analysis of WIND data shows. Insurance companies are the second largest institutional investors in A-shares. Chinese insurers have been allowed to hold equity shares on their own account since 2004. Chinese portfolio regulations allow insurance companies and their asset management arms to invest up to 30 per cent of their total assets in Ashares. As of 31 March 2020, insurance companies owned 4.10 per cent of Ashares, as measured by market capitalization. 78 Their investments at the firm level were far more focused than those of the securities investment funds. Chinese insurers were present in 228 (out of 1,980) mainboard-listed firms, with an average of 4.80 per cent (median: 2.06 per cent) of the shares in their invested companies, as shown by our analysis.

Pension funds are another important category of institutional investor. The principal player in this group is the National Social Security Fund (NSSF), which was established in 2000 by the Chinese central government to fund the country's social security system. The NSSF may invest in stocks, fixed-income products, and unlisted equities. As of 31 March 2020, the NSSF owned 2.08 per cent of A-shares, as measured by share value. 79 At the firm level, at the end of 2019, the NSSF was counted as a shareholder in 308 (out of 1,980) mainboardlisted firms, with an average of 2.58 per cent (median: 1.68 per cent) of the shares in their invested companies, according to our analysis of WIND data. Enterprise annuities, the Chinese equivalent of corporate pension plans, have only a modest presence in the Chinese stock market, and were seen in only 3 (out of 1,980) mainboard-listed companies.

Other As indicated above, the Qualified Foreign Institutional Investor (QFII) is a unique type of institutional investor in the Chinese A-share market. QFIIs are, in effect, a selected cohort of foreign institutional investors specially approved by the PRC authorities to invest, typically on behalf of their ultimate beneficiaries, in Ashare stocks. At the end of 2018, 309 QFIIs had been licensed. 82 It remains unclear whether institutional investors in China's A-share market play an active stewardship role. Anecdotal evidence suggests that institutional activism has helped to translate the principles of corporate governance from theory into actual practice, thereby raising corporate governance standards in China. 83 This view is supported by strong empirical evidence showing that higher levels of securities investment fund ownership deter management from initiating valuedecreasing transactions. 84 The same can be said about QFIIs, which have helped to significantly reduce expropriation by controlling shareholders, as some empirical research has shown. 85 On the other hand, some studies have suggested that institutional investors-particularly securities investment funds-are significantly less likely than individual investors to vote against management proposals. 86 This finding is at least partly explicable by the relatively low levels of institutional ownership in China. As indicated above, institutional investors as a group own less than 20 per cent of A-shares by value and, on average, less than 5 per cent (with even lower median values) of their portfolio firms' outstanding shares. This level of ownership stands in sharp contrast to the 'institutionalization' of the US market, 87 where institutions own 80 per cent of all stocks in the shareholder. Despite its importance, the CSISC is not the focus of this article because its shareholding, measured by its size, is almost negligible for the purposes of this research. 82 Standard and Poor's 500 (S&P 500) and one of the 'Big Three' asset management firms (that is, Blackrock, Vanguard, or State Street) is the largest shareholder in 88 per cent of all S&P 500 firms. 88 Considered globally, institutional ownership in China lies towards the lower end of the spectrum. 89 The mixed evidence on institutional stewardship is also explicable by the fact that the notion of stewardship only recently began to gain traction in China's corporate governance regime, following the overhaul of the Code of Corporate Governance in 2018. 90 Although the Code's official endorsement of the concept of institutional stewardship marked an important step forward, its approach was characteristically light-touch. Stewardship is recommended for institutional investors, but is not mandatory. The Code states that institutions are 'encouraged' to exercise their shareholder rights, including the right to vote, 91 and that they are also 'encouraged' to disclose their strategies on shareholder voting, and so on. 92 These general statements, albeit desirable, are not supplemented by bright-line rules on institutional voting, let alone provisions for legal consequences.

The hundreds of millions of Chinese retail shareholders 93 are a characteristically heterogeneous group. They differ greatly in their shareholding stakes. Chinese retail shareholders are customarily divided into sub-groups based on the value of their individual shareholdings: those holding less than RMB 100,000 (approximately US $14,000), between RMB 100,000 and RMB 500,000, between RMB 500,000 and RMB 1 million, between RMB 1 million and RMB 3 million, between RMB 3 million and RMB 10 million, and over RMB 10 million. Broadly speaking, there is an inverse relationship between a sub-group's membership and the value of the shares it represents. Accordingly, sub-groups where each member owns less than RMB 100,000 in share value account for by far the largest number of individual investors, amounting to over 55 per cent of all domestic retail investors in the Shanghai market. At the same time, these sub-groups 88 account for only 1.25 per cent of the market as measured by share value. 94 By contrast, sub-groups where each member's share value crosses the RMB 10 million mark account for less than 0.5 per cent of the individual shareholders investing in the Shanghai market, but are worth close to 7 per cent of the value of the whole market. 95 Retail shareholders can be found across a wide spread of age groups, but are heavily concentrated in the younger age group. Close to 40 per cent of individual shareholders are below the age of 30, and just over 30 per cent fall into the 30À40 age group. Less than 5 per cent are over the age of 60. 96 Individual shareholders also have widely differing education levels. Less than 25 per cent of retail shareholders are educated to degree level or above, and nearly 25 per cent do not even have a high-school level of education. 97 Individual shareholders also vary greatly with regard to investment experience. In a recent official survey of close to 200,000 retail investors, 36.1 per cent of those surveyed claimed to be in possession of sound professional knowledge. In contrast, 48.6 per cent selfreported as having a 'basic understanding [of] investment', and 15.3 per cent characterized themselves as 'novice' investors. 98 As many Chinese retail shareholders hold small stakes, are poorly educated, and lack investment experience, their profile seems to fall squarely within the conventional wisdom about 'rational apathy'. It is received wisdom that Chinese retail investors, who dominate the A-share market in terms of numbers, are driven by 'speculative motives'. 99 They account for as much as 80 per cent of equity trading, and are widely perceived to have contributed to the notorious market volatility and erratic pricing observed in the A-share market. 100 Put bluntly, small, individual shareholders are often perceived to behave more like gamblers in a casino than like value-driven investors.

However, several recent studies suggest that some retail investors take their voting rights seriously when provided with a conducive institutional environment. So, when a Chinese corporate governance reform initiative was rolled out to empower individual shareholders by giving them a veto right, they were more likely to vote against value-decreasing management proposals in firms featuring higher levels of securities investment fund ownership. 101 Online voting is another example. In earlier years, when online voting was enabled in some firms but not in others, the firms allowing online voting saw a 35 per cent higher shareholder 94 Shanghai Stock Exchange Statistics Annual 2019 (2020) 583. 95 turnout than those that did not. The number of minority shareholders participating in the former group was five times larger than in the latter group. 102

How minority shareholders vote: has it differed during the pandemic?

The regulatory regime for listed company disclosures in China offers an enabling environment for undertaking an empirical investigation into the voting behaviour of minority shareholders as a group. First, Chinese listed companies have, since 2006, been required to disclose information on shareholder turnout and voting outcomes after each shareholder meeting. Such disclosures include the number of shareholders (and their proxies) attending the shareholder meeting, the number and percentage of shares they hold, and the voting outcomes of each resolution laid before shareholders. 103 Particularly relevant to our study is an additional disclosure requirement introduced in 2014 in relation to minority votes: Chinese listed companies are now also required to disclose separately the votes cast by minority shareholders as a group on resolutions deemed to have a significant impact on their interests. The purpose of this requirement is to amplify the 'voice' of minority shareholders, thereby incentivizing their participation in the governance of their invested firms. 104 It is important to distinguish the disclosure requirement pertaining to minority votes from the majority of minority rule. The former requirement is, by nature, a disclosure rule, and does not confer upon disinterested, minority shareholders the power to veto a resolution. However, data on shareholder voting in Chinese listed companies, albeit publicly accessible due to the disclosure requirement, are not systematically available in any extant commercial database.

To remedy this data gap, we have specifically created for this research a unique, comprehensive dataset using data-mining and machine-learning techniques. We are therefore able to present the first empirical evidence on both minority shareholder voting during the Wuhan lockdown period in 2020 and the votes cast by minority shareholders during the same time periods in the five years from 2015 to 2019.

Our data on shareholder voting were extracted from the company disclosure documents released after each and every shareholder meeting of all companies listed on the main board of the Shanghai Stock Exchange during the 23 January to 7 April period in each of the six years between 2015 and 2020. The source of the sample firms' disclosure documents is CNInfo, 105 designated by the Chinese regulatory authorities for the online release of disclosure documents. The documents were in PDF format, downloadable from CNInfo, and are mostly machine readable. Data were subsequently extracted from them using data-mining and machine-learning programs developed for the specific purposes of this study. The focal time period under investigation is 23 January to 7 April. It was during this period in 2020 that Wuhan, the epicentre of the COVID-19 outbreak in China, was locked down on the orders of the Chinese government in an attempt to rein in the rapid spread of the coronavirus. Many other Chinese cities also experienced a Wuhan-style lockdown during the same period, but typically for a shorter duration. A number of other cities also experienced lockdowns after April 2020, but they were of a much smaller geographic scale. Therefore, the 76-day period from 23 January to 7 April serves as a convenient proxy for the difficult period in which the COVID-19 outbreak presented China with a severe public health and economic crisis, unprecedented in recent decades.

We start our investigation in 2015, the first year in which the requirement for the separate disclosure of minority votes, first introduced in 2014 as noted above, was universally complied with. 106 We focus on the Shanghai Stock Exchange rather than the Shenzhen Stock Exchange largely because of a unique political economy factor: in 2000, all new initial public offerings on the mainboard of the Shenzhen Stock Exchange were halted. All Chinese mainboard initial public offerings since 2000 have been floated in Shanghai, and all mainboard companies listed in Shenzhen went public more than two decades ago. Firms listed on the Shanghai Stock Exchange therefore present an unbiased picture of China's listed sector insofar as the diversity of listed companies is concerned.

Our Table 1 demonstrates a consistent pattern regarding the number of shareholder general meetings held during the sample periods (23 JanuaryÀ7 April) between 2015 and 2020. Except in 2015, 500 to 525 shareholder meetings took place each year, consistently accounting for slightly over 10 per cent of the total number of general meetings convened in the respective years. 107 The requirement to disclose minority votes separately (which became effective in 2014 and has been fully implemented since 2015) was triggered in a majority of the sample general meetings during the six sample periods, as Table 1 illustrates. 2017, and generally flattened thereafter. The upward tick in voter turnout is similar to that seen over the past decade in a number of European jurisdictions where concentrated ownership is the norm. 108 China's total shareholder turnout is, however, lower than in these jurisdictions, where it typically stands around the 60 per cent mark, if not higher. 109 Despite its importance, total shareholder turnout can be little more than a smokescreen. The largest shareholders in Chinese listed companies own an average of more than 35 per cent of their controlled firms, as mentioned above. These controlling shareholders and their close-knit associates can feature so prominently in total shareholder turnout that outside minority shareholders may be unable to exert a significant influence.

This proposition finds strong support in our data on 'minority shareholder turnout', defined as the aggregate percentage of shares held by the minority shareholders participating in the sample meetings. 110 Table 1 reports extraordinarily low minority shareholder turnouts during the periods under Ibid. 110 Chinese listed firms are not required to disclose minority shareholder turnout at shareholders meetings. They have been required, however, to disclose the votes cast by minority shareholders as a group on resolutions deemed to have a significant impact on their interests ('qualified resolutions') (see note 104 above). The minority shareholder turnout is calculated as the ratio of (i) the aggregate number of minority shareholder votes cast (including votes for, against and abstentions) on a given qualified resolution and (ii) the total number of outstanding shares of the company concerned. Two caveats are in order. First, it is possible that a shareholders meeting has no qualified resolution, in which case the minority shareholder turnout for that particular meeting is not measurable. Second, a shareholders meeting may have more than one qualified resolution and, albeit less than common, the aggregate minority shareholder votes may differ across those resolutions. In this case, it is the highest value of aggregate votes that is deemed as the numerator of the minority voter turnout.

investigation. The average minority voter turnouts moved by between only 4 and 5 percentage points, and the median values barely exceeded 3 per cent. This stands in sharp contrast to the collective holding, on average, of retail shareholders and institutional investors, which was just over 50 per cent of the outstanding shares in their invested companies (median: 45.49 per cent).

In general, each share is entitled to one vote (one share one vote, or OSOV) on each matter voted on at a shareholder meeting under PRC law, 111 subject to certain exceptions discussed below. Shareholders attending a general meeting can vote either for or against each matter laid before them, or abstain from voting. Votes for and against count towards the total votes cast, as do abstentions.

Using advanced data-mining techniques, we were able to extract from the sample company disclosures the numbers and percentages of votes cast for, votes cast against, and abstentions for each and every resolution put before shareholder meetings held during the sample periods from 2015 to 2020. Our dataset therefore includes 2,133, 3,373, 3,032, 2,749, 2,246, and 2,828 sample resolutions for the six respective 76-day sample periods, as Table 2 demonstrates. Taking the annual numbers of sample shareholder meetings reported in Table 1 into account, there were an average of 6.67 proposals per shareholder meeting, a significantly smaller number than the 13 proposals per meeting seen in Europe. 112 Before we proceed to a discussion of our empirical findings, a distinction should be noted between management proposals and shareholder proposals. 113 Shareholder proposals are a hallmark of the rising shareholder activism seen particularly in the USA, and have been illuminated by a large body of academic and professional studies. 114 Shareholder proposals are, however, much less prevalent in the European context. 115 As far as China is concerned, we are still developing a machine learning program that can identify, systematically and reliably, shareholder proposals in the Chinese listed sector. 116 116 We attempted an identification strategy that involved examining the resolutions receiving low levels of shareholder approval, but with only limited success. High degrees of shareholder dissent (less than 1 per cent of all resolutions randomly sampled). Therefore, for the purposes of this study, we treat all sample resolutions as management proposalsthat is, proposals initiated and backed by the company's management or controlling shareholder. Given the large size of our samples (over 16,000 resolutions) and the low probability (estimated to be lower than 1 per cent) of shareholder proposals, we are confident that the broad conclusions drawn from our dataset are robust and reliable. 117 We start with the overall shareholder approval rate at the resolution level, defined as the total number of votes cast in favour of a sample resolution divided by the total votes cast on that resolution. 118 As Table 2 are a useful, but not always reliable, indicator of the proposals concerned being shareholder proposals. 117 As a matter of statistical technique, winsorization of our data could perhaps eliminate some shareholder proposals. However, as indicated above, (low) approval rates are not a reliable indicator of shareholder proposals (or outliers). Winsorization is therefore not a reliable method of censoring those proposals. 118 The shareholder approval rate is calculated as the ratio of (i) the total number of votes for and (ii) the total number of votes cast on a resolution. As noted above, the votes for and against and abstentions add together to become the denominator. European jurisdictions (including the UK), which recorded a mean of 96.3 per cent and median of 99.3 per cent. 120 We further break down the sample proposals into several groups based on their approval rates: proposals receiving unanimous shareholder support; proposals supported by more than 99.9 per cent but less than 100 per cent of votes cast; proposals whose approval rates fall into the 99À99.9 per cent range, into the 90À99 per cent range, and into the 50À90 per cent range; and proposals favoured by less than 50 per cent of votes cast. The 99.9À100 per cent and 99À99.9 per cent brackets may, at first sight, seem excessively narrow, but make sense in the light of the high mean and median rates discussed immediately above. 121 The figures reported in Table 2 are strikingly 'front-loaded' in that the sample proposals clearly cluster in the 100 per cent and 99.9À100 per cent groups. The other end of the spectrum-that is, the 0À50 per cent bracket-is extraordinarily thin.

Specifically, resolutions supported unanimously by attending shareholders accounted for as much as 40.79 per cent of all sample resolutions in 2015. 122 Unanimous shareholder support became much less frequent in the following years, dropping sharply to approximately 15 per cent of all resolutions in the 2016 and 2020 sample periods. However, not too much should be made of this, as it takes just one dissenting shareholder holding as few as 100 shares to move a proposal down into the 99.9À100 per cent bracket. When the 100 per cent and 99.9À100 per cent groups are lumped together, it becomes immediately apparent that unanimous and close-to-unanimous shareholder support for management proposals constitutes the norm in China. Such proposals consistently amounted to over 60 per cent of all resolutions across the six sample years, and in 2018 and 2019 they exceeded 70 per cent.

In sharp contrast, the group of sample resolutions falling into the 0À50 per cent bracket was almost negligible. 123 They accounted for only 0.1 per cent of the total number of sample resolutions in 2017, and barely crossed the 1.5 per cent threshold in 2016, as Table 2 illustrates. It might seem reasonable to suppose that the few proposals that fell into this category were likely shareholder proposals, which are, as comparative studies suggest, much less likely than management proposals to succeed. 124 Closer scrutiny does not, however, support this proposition, as many management proposals found their way into what appears to be an unlikely destination. This phenomenon is beyond the scope of this study, but offers fertile ground for future case studies of shareholder activism in China.

Voting patterns of minority shareholders: did the pandemic have an impact?

The overall patterns of shareholder voting in Chinese listed companies depicted above seem to paint a rosy picture: in the years under study, management proposals in China generally met with little (minority) shareholder opposition. This picture may, however, be misleading. Low levels of minority participation in shareholder meetings can distort the voting outcome in favour of management and/or controlling shareholders, 125 particularly in China. The presence of a majority shareholder can easily crowd out minority dissent. It is therefore important to delve deeper and take a more focused look at how minority shareholders vote, so that a more nuanced understanding of their preferences and behavioural patterns in respect of voting can be obtained.

Shareholders can vote strategically, 126 and their voting practices can differ by proposal type. 127 Recent empirical research suggests that Chinese shareholders also seem to distinguish between proposals more susceptible to conflicts of interest and those that involve little conflict in casting their votes. 128 Accordingly, we categorize our sample resolutions into the 13 following types, using the classification framework prescribed by the PRC Company Law and the CSRC-enacted Model Articles of Association, 129 a framework that is intended for 'guidance' purposes but is generally followed by Chinese listed companies: 130 • annual budgets and accounts; 133 • profit distribution and loss coverage; 134 • capital increases or decreases; 135 • bond issuance; 136 • mergers, splits, dissolutions, or liquidations; 137 • charter amendments; 138 • appointments of accountants; 139 • guarantees; 140 • substantial acquisitions or disposals of assets; 141 • share-option schemes; 142 and • other.

It can be seen from the foregoing list that Chinese shareholder meetings consider a wider and more diverse range of matters than do their counterparts in, for instance, the UK. 143 This diversity is beyond the scope of this research, but suffice it to say here that it is attributable to China's shareholder-primacy approach to the division of power between shareholders and boards. 144 In China, powers are conferred upon shareholder meetings that, in many other jurisdictions, are vested in the board. 145 Two important caveats are in order before we proceed to examine the voting patterns of Chinese minority shareholders, both of which are exceptions to the general principle of OSOV. 146 First, resolutions on director and supervisor elections are excluded for the purposes of this study, largely because of the common use of the cumulative voting rule in regard to the election of Chinese dual-board members. 147 Second, the relatively small cohort of companies listed on the nascent Shanghai-based STAR market, which is, in effect, China's Nasdaq-style tech board, is also excluded because dual-class shares are permissible for those firms. 148 'Minority voter approval rates' are defined as the total number of minority votes cast in support of a proposal divided by the total number of votes cast on that proposal by all attending minority shareholders. Figure 1 shows the minority approval rates in respect of 12 types of resolutions (excluding 'Other') over the six sample periods. It is now clear that the overall approval rates reported in Table 2 downplay the opposition of Chinese minority shareholders. It was rare for the minority approval rates to approach 96.93 per cent, the lowest overall shareholder approval rate (recorded in 2016). 149 Votes against and abstentions often accounted for over 10 per cent of the total minority votes cast, and they sometimes went as high as 70 per cent (bond issuances in 2015). Minority shareholder disapproval beyond the 20 per cent level was not uncommon (for Note that the annual distributions of resolution-level approval rates are placed at varying scales across the resolution types. In the brackets are the annual numbers of sample resolutions (Ns) for each resolution types. board work reports in 2019 and 2020, annual budget and accounts in 2020, and share-option schemes in 2016).

More importantly, the time series data presented in Figure 1 lend strong support to the proposition that shareholder voting behaviour varies across the 12 resolution types, with the annual minority approval rates falling into different ranges depending on the type. Counterintuitively, however, a closer look at Figure 1 shows that routine resolutions seem to receive more votes against and abstentions than do those likely to have pronounced shareholder wealth effects. Hence, substantial acquisitions or disposals of assets, which are widely believed to offer controlling shareholders an opportunity to tunnel the controlled listed companies in question, recorded approval rates above 90 per cent in four of the six sample periods. By contrast, the annual work reports of boards of directors and supervisory boards, which ought to have been relatively uncontroversial items of business, never managed to cross the 90 per cent mark in any of the six sample periods and, indeed, in two of the sample periods even fell below the 80 per cent mark. This counter-intuitive observation, particularly in respect of the 2020 sample period, will be discussed in greater detail in the next part. Suffice it to say here that Chinese minority shareholders, insofar as they vote, do not seem to fit well into the conventional stereotype of indifferent/apathetic voters. Instead, they seem to direct a lot more of their attention to certain types of management proposals, thereby playing a visible monitoring role. This finding adds to the emerging empirical evidence suggesting that minority (retail) shareholders do utilize their voting power as a means to monitor their invested firms. 150 Our central line of inquiry is, of course, whether Chinese minority shareholders demonstrated a discernibly different voting pattern under the impact of the COVID-19 pandemic. It should first be noted that the minority approval rates fluctuated over the six sample periods, and that these fluctuations were sometimes significant in respect of certain resolution types, as Figure 1 illustrates. For example, minority voter support for business policies and investment plans remained relatively stable over time, moving within the relatively narrow range of 89.04 per cent to 92.13 per cent. By contrast, opposition to bond issuances swung wildly, from as much as 69.62 per cent (in 2015) to as little as 0.23 per cent (in 2018). These fluctuations do not seem to have been caused by outliers, as our sample sizes tend to be sufficiently large overall to smooth out random 'noise'. 151 Upon a closer look, however, a distinct pattern of minority shareholder voting emerges in the 2020 sample period. Minority approval rates during the pandemic period often tilted towards either the higher or lower end of the spectrum across the six sample periods, depending on the proposal type. Specifically, minority voter dissent reached its highest level in 2020 in respect of four proposal 150 See eg note 14 above. 151 The only exception is the proposal type of mergers, splits, dissolution or liquidation. These are fundamental changes to listed companies, and they occurred with significantly lower frequency than other resolution types. Despite their small sample sizes, this proposal type's approval rates do not swing particularly widely (81.50À91.99 per cent).

types: business policies and investment plans; annual work reports of (dual) boards; annual budgets and accounts; and profit distribution and loss coverage. By contrast, during the 2020 sample period minority shareholders voted most frequently in favour of management in respect of two proposal types: guarantees and substantial acquisitions/disposals of assets. Minority voter support reached its second-highest level in 2020 in respect of two other proposal types: capital increases/decreases and share-option schemes. So, relative to the five other sample periods, the 2020 sample period demonstrated by far the greatest degree of clustering and concentration at both ends of the spectrum. Moreover, the counterintuitive pattern of minority approval rates, as noted above, became more apparent and pronounced during the pandemic period: routine matters met with the highest level of minority opposition, whereas potentially conflict-ridden proposals received the highest percentage of 'for' votes from minority shareholders. The unique pattern of voting outcomes is a strong sign of Chinese minority shareholders voting in a more considered and selective, if not more intelligent, manner in the 2020 sample period. To be sure, it could be argued that Chinese minority investors-who are known to be speculative-might not, after all, have been as informed as our analysis suggests that they were. In the period in question, they may instead have been simply (over-)reacting to a once-in-a-century public health crisis and the ensuing economic and market downturns. This alternative theory would be more credible if our data had not shown minority voters to have voted selectively during the 2020 sample period. Rather than all types of proposals having received more for or against votes across the board in that period, some proposal types recorded the highest levels of shareholder approval, and others the highest levels of shareholder dissent. Yet another theory might be that the sample firms in 2020 'selfselected' themselves, and therefore presented a year cohort bearing character that differed fundamentally from those of the cohorts in the five other sample periods. However, the (unreported) robustness check we conducted in respect of such firm-level attributes as ownership structure and institutional shareholding revealed no evidence of self-selection bias. The remarkable patterns observed during the pandemic in 2020 demand an explanation, to which we now turn.

The pandemic pattern of minority shareholder voting: what changed rational passivity?

The received wisdom in corporate law scholarship is that collective action problems can lead to rational shareholder passivity. 152 For shareholders, active monitoring comes at a cost that is borne largely, if not entirely, by themselves, whereas the benefits of monitoring are gained on only a pro rata basis. Shareholders are economically incentivized to monitor and engage only when the potential gains of doing so are greater than the costs. 153 In other words, a shareholder's incentive to actively monitor is supposedly determined by a trade-off between the private costs of monitoring, which are fully internalized by the shareholder, and the public benefits of monitoring, which are shared among all shareholders. 154 As noted in the preceding part, Chinese minority shareholders seem to have voted in a more considered, if not more intelligent manner, than usual during the 2020 sample period. Specifically, they more frequently cast votes against certain types of management proposals or registered abstentions, and voted more often in favour of other proposals, than in the previous sample periods. Viewed through a cost-and-benefits lens, the observed change in minority voter behaviour during the pandemic may well be attributable to a shift in the pre-existing cost-and-benefits trade-off. Presumably, such a shift was induced by either increased potential benefits or reduced costs, or both.

We start with the costs of minority shareholders voting more intelligently. An important source of such costs is the time and resources allocated to evaluate the merits of individual management proposals and, where necessary, to acquire further company-specific information to enable minority shareholders to cast an informed vote. For institutional investors, dedicating time to voting and stewardship at a particular investee company can be costly. 155 Such costs, however, became considerably lower during the COVID-19 pandemic in China due to the widespread use of state-ofthe-art virtual technologies. Take 'corporate visits', a particular stewardship method by which Chinese institutional investors monitor their portfolio companies, for example. On these visits, fund managers, analysts, and other personnel of institutional investors typically hold face-to-face meetings with management. Empirical research shows that such visits can serve an important governance role, helping to improve information disclosure 156 and to discipline a company's management. 157 Corporate visits are, however, costly in terms of the time invested in preparing for and conducting them. 158 During the COVID-19 pandemic, traditional face-to-face business interactions were generally replaced by virtual meetings enabled by such enterprise communication platforms as Alibaba's DingTalk, whose number of monthly active users rose from approximately 65 million at the end of 2019 to roughly 177 million in March 2020, according to a McKinsey report. 159 The use of remote technologies was particularly prevalent in the financial and investment sectors. 160 On-site, physical corporate visits were therefore mostly replaced by virtual visits conducted with the assistance of modern meeting technologies. 161 The adoption of remote technologies would significantly reduce the time and costs of travelling in a large country such as China, helping to reduce the overall costs of becoming informed and voting more intelligently. This proposition finds support in the data on corporate visits drawn from WIND. Site visits started to become popular among institutional investors in 2017. In the 2017, 2018, and 2019 sample periods, 704, 934, and 864 institutional investors, respectively, conducted corporate visits, almost invariably in the form of physical, on-site visits. By contrast, the 2020 sample period recorded 1,852 visits, more than double the average number (834 visits) in the three previous sample periods; but the overwhelming majority of these visits were conducted via video conferencing, as the WIND data show. Remote corporate visits seem to have significantly reduced the costs of acquiring firm-specific information for Chinese institutional investors, enabling them to vote in a more welfare-enhancing way. They also helped to free up time and resources that could be allocated to the monitoring of a wider range of portfolio companies, 162 yielding important benefits across the board.

The same can perhaps also be said of active retail investors. Contrary to common belief, retail investors, as owners, do have skin in the game: 'Even if a retail investor's stake in a voting outcome is relatively small, the underlying investment is likely to be economically meaningful to that shareholder'. 163 So, when they do vote, their voting decisions may well be informed, if perhaps less than sophisticated. 164 Recent empirical research suggests that retail shareholders constitute an influential voting bloc and that they are prepared to use their voting power as a means to monitor invested firms. 165 Evidence has also emerged recently that suggests that retail investors in China take their monitoring role seriously, actively engaging with their invested companies. 166 To be sure, retail investors make corporate visits far less often than do institutional investors. Hence, the cost-saving benefit of conducting corporate visits virtually, as analysed above, is of less relevance to Chinese retail investors than to their institutional counterparts.

However, there may have been a distinct source of cost reduction-that is, opportunity costs-for retail shareholders during the COVID-19 pandemic. Most Chinese retail investors are not professional investors, as indicated earlier in this article. Their other professional commitments, such as running a business or working a full-time job, compete for their time and attention. 167 So, in normal circumstances, an informed vote can come with a significant opportunity cost for retail investors-namely, the forgone benefit that they would have derived from pursuing a business interest or being dedicated to their jobs. The COVID-19 pandemic may have reduced these opportunity costs. This possibility needs to be understood in light of the social distancing policies and measures that the Chinese government introduced in order to rein in the transmission of the coronavirus. People were encouraged to stay at home, social contacts were limited, government offices were closed, and only limited segments of the public transport system (including commercial flights) remained in operation. 168 According to a widely cited report issued by Baidu, the Chinese equivalent of Google, nationwide travel within China declined year-on-year by 58 per cent during the 22 JanuaryÀ9 February period of 2020, and did not return to normal until 31 March. 169 A UBS report further suggests that the recovery was slower than the Baidu report indicated, and that travel was at a level 20 per cent lower than in the previous year in early May 2020. 170 Therefore, business travel and meetings would have been severely curtailed during the 2020 sample period. In any event, the sharp slowdown in the Chinese economy during the pandemic meant that there were fewer business and professional opportunities to be pursued. Chinese retail investors would have chosen to use at least some of the extra time on their hands to learn more about their invested companies. These propositions found some support in the reported changes in retail investor behaviour during the lockdown. In Wuhan, private entrepreneurs found themselves awash with free cash flow (due to temporary business and factory closures), and they reportedly chose to stay at home investing the cash flow on their A-share portfolios. Small retail shareholders in Wuhan were also reported to have more time to 'study and invest in shares'. 171 Across the country, a survey of 10,000 Chinese fund investors showed that 68 percent of the respondents 'devoted more time and efforts on [understanding] their finance and investments'. 172 These retail investors would have found the opportunity cost of becoming informed to be considerably lower than it would have been in the previous sample periods (when the pursuit of business or professional opportunities was not hampered by a pandemic).

We have shown that the costs of an informed or considered vote may well have been lower during the 2020 sample period. We now proceed to examine the potential benefits for minority shareholders of voting intelligently during that period. One important consideration here is the size of the stakes held by minority shareholders. All else being equal, the larger a shareholder's share ownership is, the greater the potential gains he or she can reap from becoming informed and active in voting and engagement. In other words, a larger stake provides a greater financial incentive for an investor to evaluate the merits of management proposals and to act on the knowledge gained by voting in a value-enhancing way. 173 Shareholding data, however, are not generally available at the individual shareholder level, as minority shareholders are not required to disclose their holdings unless they are among a firm's top-10 shareholders by shareholding.

An alternative approach is to look at the aggregate shareholding by all attending minority shareholders at the firm level, as measured by minority turnout. 174 If the increased monitoring by minority shareholders observed in the 2020 sample period was driven by larger stakes held aggregately by minority shareholders (thereby justifying the greater costs of casting intelligent votes) in that period relative to previous sample periods, we would have witnessed greater minority voter turnout in the 2020 period. But this is not what our data suggest: the overall minority turnout in the 2020 sample period, as can be seen in Table 1 , did not differ appreciably from that in the five previous periods. It stood at 4.45 per cent in the 2020 period, slightly lower than in 2018 (4.90 per cent) but of the same order of magnitude as in 2015 (4.43 per cent) and 2017 (4.41 per cent). Hence, the size of minority voters' stakes does not adequately explain the significant change in minority voter behaviour seen in the 2020 sample period.

More active engagement in voting by minority shareholders can also bring about benefits if management proposals become more value-enhancing as a consequence. Related-party transactions are of particular relevance here. During the 2020 sample period, minority shareholders reportedly took a characteristically activist approach in respect of related-party transactions. For instance, on two days in a row in February 2020, management proposals to reduce the pricing of private placements (citing the pandemic-induced market downturn) were vetoed by dissenting non-interested minority voters in two Chinese listed companies. 175 In both cases, the companies' majority shareholders were themselves among the placees, and the majority of minority rule therefore applied to vest veto power in minority shareholders. As a result, there appeared to be a real threat that a management proposal featuring terms perceived as unfair could be vetoed by informed, active minority shareholders. To avoid such a shareholder revolt, the management would have been well advised to ensure that the terms of any proposed related-party transaction were perceived as fair and as enhancing shareholder welfare. The pressure on management would have been particularly acute during the 2020 sample period, as engagement in their portfolio companies by Chinese institutional investors (by way of corporate visits) intensified in that period, as discussed above. As such, investments in acquiring firmspecific information and engaging management would have stood a better chance of being rewarded with real benefits for minority shareholders, and related-party transactions would have become more value-enhancing than in normal circumstances. The possibility of such a reward would, in turn, have incentivized minority shareholders to engage with their invested companies either publicly (via, for instance, corporate visits) or privately. 176 If this engagement had been successful in optimizing management proposals for related-party transactions to the benefit of minority shareholders, minority voters would have been more willing to vote in support of those transactions than they otherwise would have been. This proposition does, indeed, find support in our data. As seen above, one of the counterintuitive empirical findings in our research is that minority voter dissent did not rise in the 2020 sample period in respect of proposals known to be susceptible to conflicts, such as guarantees, substantial acquisitions and disposals of assets, capitalization, and share-option schemes. 177 To the contrary, the 2020 sample period recorded the highest (or second-highest) levels of minority approval in respect of these four types of management proposals.

The dynamics of related-party transactions also seem to have had a spill-over effect on some of the more routine, mundane types of proposals. As noted above, minority voter dissent was at its highest level in 2020 among the six sample periods in respect of the annual work reports of a company's (dual) boards, annual budgets and accounts, and profit distribution and loss coverage. This counterintuitive finding appears to be consistent with our analysis above: minority shareholders, particularly institutional investors, motivated by the reduced costs and increased benefits of monitoring during the COVID-19 pandemic, may well have dedicated more time and attention than usual to all management proposals put to a vote at shareholder meetings. Of course, minority voters do not enjoy veto power over such routine matters as management performance, company accounts, and dividend payouts, as they do in respect of related-party transactions. However, as a spillover effect of their more diligent scrutiny of related-party transactions, these voters probably gave more time and attention to routine proposals than they otherwise would have, and voted in a more informed manner. Although they were unable to prevent the proposals from passing, they were at least able to signal to the management their disapproval of how routine corporate matters were managed.

Our analysis above reveals how the COVID-19 pandemic in China created a rare set of conditions that shifted the cost-and-benefits equilibrium of shareholder voting. Minority voters appear to have cast their votes in a more informed manner, likely enhancing their own welfare, during the 2020 sample period. 

Search of "Absent" Shareholders: A New Solution to Retail Investors' Apathy

Do Shareholders Vote Strategically? Voting Behavior, Proposal Screening, and Majority Rules

See eg note 11 above

Shareholder Voting and Corporate Sustainability in China: An Empirical Study

Promulgated by the CSRC, the Model AoA have been in force since

An Empirical Study of Corporate Default Rules and Menus in China, Hong Kong, and Taiwan'

Chinese listed companies are generally required by law to put in place a dual board structure, featuring a board of directors and a supervisory board. More generally, see Chao Xi

The Costs of Shareholder Activism: Evidence from a Sequential Decision Model' (2013) 107

Index Funds and the Future of Corporate Governance: Theory, Evidence, and Policy

The Governance Role of Institutional Investors in Information Disclosure: Evidence from Institutional Investors' Corporate Visits

Soft Activism and Corporate Dividend Policy: Evidence from Institutional Investors Site Visits

A financial analyst would be placed under considerable physical stress if she had to effectively visit more than two portfolio firms in a week

Fast Forward China: How COVID-19 Is Accelerating 5 Key Trends Shaping the Chinese Economy (McKinsey & Company 2020)

Coronavirus Outbreak Gives Rise to More Complex Forms of Remote Work

Telephone interviews with PRC fund managers: interviews 2020À4

See note 14 above

Do You Hear the People's Saying?: The Voice of Individual Investors

We assume that more resourceful retail investors, particularly more sophisticated and experienced investors, are more likely to vote. For empirical evidence of this, see eg André Schmidt, 'Determinants of Corporate Voting: Evidence from a Large Survey of German Retail Investors

COVID-19 Control in China during Mass Population Movements at New Year' Lancet

How Can We Track the Recovery in China' UBS

Traded Shares While Fighting the Pandemic: Their Willingness to Invest Topped Chinese Major Cities

Has Your Investment Behaviour Changed over the Pandemic ?' China Economic Daily

Drawing upon an original dataset showing how shareholders cast their votes in Chinese listed companies during the 2015À20 period, this study, for the first time, presents systematic evidence of Chinese minority shareholders, in defiance of the apathetic investor stereotype, disapproving management proposals, sometimes in large percentages. The research highlights the potential of shareholder empowerment in economies in which it seems unlikely to prosper-that is, economies characterized by concentrated share ownership, dominance of the market by small and unsophisticated retail investors, and a relatively weak institutional investor base.This study also enriches the emerging literature on corporate law and the COVID-19 pandemic. 178 The case of China offers a fresh perspective. Unlike in many other jurisdictions, the pandemic had little impact on China's pre-pandemic corporate law, particularly its law on shareholder meetings and voting. However, Chinese minority shareholders demonstrated a more considered, if not more intelligent, pattern of voting during the period of the Wuhan lockdown in 2020 than in the same time windows of the five-year period from 2015 to 2019. Counterintuitively, they voted more frequently in favour of types of proposals known to be more susceptible to conflicts than others, on the one hand, and, on the other, registered more often their dissent against other types that are routine in nature. This article shows that this welcome shift in voting patterns is attributable, at least in part, to the widespread use during the pandemic of state-of-the-art remote technologies, which seems to have considerably lowered the cost for minority shareholders, institutional investors in particular, to monitor their invested firms. If these exceptional measures remain in place after the pandemic has run its course, they promise to shift the cost-and-benefits equilibrium of shareholder voting and engagement and to blunt the force of shareholder apathy. 178