Towards a Virtual General Meeting : ‘ I accept ’ or ‘ I decline ’ ?

email: wanne@pemmelaar.com. This article is a revised version of the bachelor’s dissertation by Wanne M. Pemmelaar, entitled ‘Naar een virtuele aandeelhoudersvergadering: een vrije val of een bijdrage aan corporate governance?’ that was completed on January 28 2008 under the supervision of Sanne-Roos van der Waals LLM. at the Molengraaff Institute of Private Law of the Faculty of Law, Economics & Governance, Utrecht University, the Netherlands. 1 Hans Pronk is an Internet consultant and currently a partner at the consultancy firm Verdonck, Klooster & Associates based in Zoetermeer (the Netherlands). 2 A. van der Krans,. ‘De virtuele aandeelhoudersvergadering’, 2007 Ondernemingsrecht, no. 7, p. 277. 3 A. de Jong et al., The Role of Self-Regulation in Corporate Governance: Evidence from The Netherlands, 2001, www.erim.eur.nl (accessed 11 November 2007), p. 1. 4 Ibid., p. 1.


Introduction
At a 'diner pensant', recently organised by the Dutch law firm Stibbe, the Internet expert Hans Pronk announced that 'the future is already here (…) it is just not widely distributed yet'. 1 The Internet has currently moved past its scarcity.The possibilities regarding modern communication technologies are already at hand.However, the Dutch (and EU) legislature is no entrepreneur in acknowledging new media technologies as a means to validly conduct legal transactions.The Dutch legislature will nonetheless be increasingly confronted with requests to permit more technological provisions in the Burgerlijk Wetboek (the Dutch Civil Code, hereafter abbreviated as 'BW').One issue that will require more attention is the use of modern communication technology during shareholders' meetings.One possibility is the use of the so-called Virtual Shareholder Meeting (also known as the Virtual General Meeting, 'VGM').This will form the centrepiece of this article.
Over the course of the last few decades there has been wide interest among the legislature, academics and practitioners for shareholders' rights and the exercise of these rights. 2De Jong et al. state that due to increased international competition and the integration of Europe the focus has shifted towards companies and industries that are restructuring in order to cope with competition and to promote economic growth. 3Since conventional sources of capital are limited, attention has focused on capital markets.Corporate governance plays a crucial role in capital markets in order to determine where, in what form, and at what cost capital is provided by external investors. 4The Dutch legislature shares the same goal as legal practitioners and academics to create a better system of checks and balances, with special interest being given to the role which the shareholder plays in this regard.Creating greater involvement among foreign 5 Arts.2:108 and 2:218 BW. 6 Although I am a proponent of the fundamental comparability of all legal systems (see: K. Zweigert & H. Kötz, Introduction to Comparative Law, 1998, p. 42), I have made a practical selection based on the aim and object of this research (see: Th.M. de Boer, 'Vergelijkenderwijs: de inspiratie van buitenlands recht', 1992 WPNR, p. 48).As the object of this research is the VGM, I have chosen two legal systems that have introduced this legal construction in their legislation.Since the aim of the research is to evaluate the VGM independently and to assess the possibilities to reform Dutch law accordingly, it would not be of any value to include other legal systems that do not include any provisions on the VGM. and small shareholders in the corporation is considered to be the key to a better functioning system of corporate governance.However, Van der Krans observes several disincentives that obstruct any increase in the participation and involvement of shareholders.The main disincentives include transnational voting, rational apathy, asymmetrical information and the unfortunate planning of the Annual General Meetings, which all take place every spring. 5t is believed that the Virtual General Meeting could contribute to swift, efficient and, above all, inexpensive communication between the corporation and its shareholders, and thus remove the disincentive of geographical distance.Delaware (US) and Denmark are the only states to date that have introduced the Virtual General Meeting in their legislation.To measure whether the introduction of the VGM in Dutch business law would be feasible and desirable it is useful to examine the functional solutions and the experience in these legal systems.This could be done on the basis of comparative legal research. 6However, considering the rapid developments in the field of IT and the reluctance on the part of the legislature to follow every trend, one needs more than comparative law alone to examine the functionality of a VGM in relation to corporate governance.An important criterion to measure whether the VGM contributes to corporate governance could be by comparing the attendance rates for Annual Shareholders' Meetings.
The nature of corporate governance and IT & Law persuades the researcher to adopt a multidisciplinary approach.This article is meant to produce some ideas on how the VGM could contribute to corporate governance, the position of the shareholder and to provide some means to measure these contributions.The first section will consider the importance of a high rate of participation by the shareholder and examine the actual attendance rate of shareholders at the Algemene Vergadering van Aandeelhouders (Annual General Meeting of Shareholders, hereafter abbreviated as 'AVA').The concept of the VGM as advanced in the academic literature will be covered in the second section.In the third section a brief glimpse will be taken at the regulation and its application regarding the VGM in Delaware (US).In the fourth section the VGM will be evaluated in the light of corporate governance and ideas on measurements will be put forward.The article will be concluded in the fifth chapter with a proposition for two VGM models.

The significance of a high rate of involvement
Good corporate governance leads to the durable existence of a corporation, which reduces investment risks. 7Lehmann and Weigand state that corporate governance instruments could, in certain circumstances, result in better performance by the corporation since the shareholder also demands high returns on his investments in the long term. 8The Dutch system of self-regulation requires an effective structure of checks and balances between shareholders and management.The effectiveness of this structure is dependent on the level of shareholder involvement concern-ing the lock, stock and barrel of the corporation.It all revolves around the attendance at the shareholders' meeting, especially since the shareholder can exercise most of his monitoring rights during and preceding the shareholders' meeting (Dutch Corporate Governance Code Principle IV.1).The board member could more likely be impressed as a higher percentage of shareholders hold him accountable. 9The approval or dismissal of a resolution by the AVA will provide more evidence of an Unternehmensgemeinschaftsgeist when the attendance rate is high. 10Another reason, according to the Frijns Committee, is that companies that have a high attendance rate run less risk of being confronted with 'chance majorities'. 11he supply of information is, besides effective monitoring, an essential element of a good functioning system of corporate governance. 12Corporate governance is well served when the shareholder shows involvement in gathering information preceding, during and following the AVA.This leads irreversibly to well-informed resolutions.

Actual attendance and involvement
A big concern, year by year, has been that the attendance of shareholders at the AVA is quite low. 13An investigation by Abma on the attendance of shareholders in 2006 at general meetings of AEX-listed companies showed a rate of 36.4% (disregarding companies that have issued depositary receipts for shares). 14The overall average attendance rate at AVAs was higher, namely 44%.The Frijns Committee, which ensures the application of the Tabaksblat Code (the Dutch Corporate Governance Code), observed in 2005 and 2006 that the shareholder shows more involvement. 15This is confirmed in recent studies in 2007 conducted by Rematch and the Frijns Committee, which showed an increase in the attendance rate for AEX-listed companies from 36.4% to 45% and the average attendance rate from 44% to 50%. 16Eumedion stated that this raise could be explained by the fact that companies published their agendas earlier before the meeting and the fact that the registration terms were brought forward. 17Although these reasons are still to be confirmed by other researches they do provide some ground for an alternative hypothesis concerning variables that could result in an increase in attendance rates.This issue will be further analysed in Section 4.2.
Although there has been a modest rise in attendance rates over the past two to three years it is too soon to call it a trend.Furthermore, it is important to state that there is still room for improvement.Attendance rates in the Netherlands are quite low compared to other countries, such as, for example, the US (85% over the last decade) and Belgium (57% over the last decade). 18

Reasons for the low involvement
Van der Krans points to transnational voting, rational apathy, asymmetrical information and the unfortunate planning of the AVAs as disincentives that obstruct any increase in the participation and involvement of shareholders. 19Although the possible causes enumerated by Van der Krans are quite plausible, it remains unclear what criteria and methods were used to establish a causal relationship between the aforementioned independent variables and the low involvement of shareholders.This raises the question whether alternative causes and hypotheses have been properly investigated.It would be wise to further explore the causes of the low involvement of shareholders, before trying to elaborate any solutions.However, with a view to the ongoing internationalisation of Dutch trade and industry, one of the most apparent causes for low attendance at AVAs seems to be transnational voting.The percentage of foreign investors increased during the period 1995-2003 from 37% to 75% while the percentage of Dutch investors (state, institutional and private) decreased over the same period from 62% to 18%. 20It seems plausible to assume that it could in fact be impossible for large (foreign) investors to attend every shareholder meeting organised during April-May of every year.However, one could also argue that it is the core business of large investors to be present and that different representatives could be sent to different AVAs.It might be a good idea to look at the (smaller) private investor.The costs of time and travel might not be in balance with the small gains for the private investor.A private investor might not believe that he has any significant impact on the course and outcome of an AVA. 21symmetrical information seems to be another apparent reason for low shareholder involvement.A survey conducted by the Frijns Committee in 2006 shows that the supply of information forms an obstacle to the involvement of shareholders. 22One could observe a gap between the way corporations and shareholders perceive the supply of information.Dutch corporations believe that they supply sufficient information to the shareholders, while shareholders indicate that they prefer to receive more information. 23This gap can partly be explained by the fact that corporations state that they prefer not to supply information on strategic decisions and the fact that they usually do not check whether the information does indeed reach the individual shareholder.Corporations also believe that they provide enough information outside shareholders' meetings.It is remarkable that the survey shows that investors do not necessarily agree with this.Very little is being done by corporations to allow shareholders to contact each other with trade information. 24he Frijns Committee stated in 2005 that the low attendance rates were due to the low accessibility of the AVA. 25 According to the Committee this could be improved by the use of electronic means of communication.Also Abma was of the opinion that, with a view to the

Introduction
The European Commission has also placed emphasis on the improvement of shareholders' rights.This resulted in the consultation report 'Fostering an appropriate regime for shareholders' rights' published on September 16 th 2004. 27The consultation report also highlights the need for improvement in the fields of transnational voting, the supply of information to shareholders and the way in which shareholders' rights are exercised.Based on this consultation report and the recommendations of the Tabaksblat Committee the Dutch legislature adopted the Wet Elektronische Communicatiemiddelen (Electronic Means of Communication Act, hereafter abbreviated as 'Wec'), which came into force on January 1 st 2007.As the Wec precedes Directive 2007/36/EC on the exercise of certain rights of shareholders in listed companies, the Dutch legislature is now in consultation with the EC to see whether the Act is in compliance with the directive. 28The Dutch Government has stated that it will evaluate practical experiences so far in order to assess which additional amendments need to be made to the law.This section will begin by providing an overview of the use of ICT before and during the AVA by evaluating the law as it now applies and the practical experiences so far (Section 2.2).After this, the ideal concept of the VGM, as put forward in the academic literature will be dealt with (Section 2.3).

Legal possibilities
The purpose of the Wec is to facilitate the use of ICT applications before and during the AVA. 29owever, the general principle of the Wec is that there is no obligation to use ICT.The AVA can now decide to make use of the following ICT applications: 1.
Already since September 2004 the law states that the requirement that the proxy must be in written form is fulfilled when this is recorded electronically (Article 2:117/2:227 §5 BW).No special statutory provision is required for exercising this right.Nowak even states that articles of association should not be able to exclude the use of electronic proxies. 302.
Another option is an electronic notice convening a shareholders' meeting, which needs no special statutory provision but can be excluded by the articles of association (Articles 2:113 §4 and 2:223 §3 BW).The corporation can now, besides the use of conventional means, notify shareholders by email (which is due to the prior consent of the recipient).The electronic notice needs to be readable and reproducible.A corporation can also publish the notice on its website, for which a statutory basis is required since shareholders need to give their prior consent jointly in this case.

3.
Every person who is entitled to attend the AVA has the right to exercise this by means of electronic communication (Articles 2:117a/2:227a BW).The basic principle is however still a meeting held at a physical place in the Netherlands; it is therefore not possible to hold an entirely virtual shareholders' meeting.According to the explanatory notes on the Wec the shareholder needs to safeguard that he can virtually attend the AVA 'as if he were present at the physical meeting place'.However, the Dutch legislature stated in its explanatory notes that with this phrase mainly audiovisual broadcasting was meant. 31One should note that this provision could lead to some practical obstacles; being able to follow the proceedings and voting does not yet mean 'as if he were present', the corporation does need to provide some way for the shareholder to digitally 'speak out'.A statutory basis is required for this application.4.
It is possible electronically to vote prior to the meeting (Articles 2:117b/2:227b BW).Due to this amendment the actual adoption of resolutions shifts towards the phase preceding the AVA.This will probably increase the demand for an earlier supply of information by the corporation.A statutory basis is required for this application.5.
It is also possible to electronically vote outside the shareholders' meeting (Articles 2:128/2:238 BW). 32A statutory basis is not required for this ICT application.However, the articles of association can exclude electronic voting outside the shareholders' meeting.

Actual use
Quantitative research, conducted by the Frijns Committee prior to the entry into force of the Wec, shows that Dutch corporations have not made much use of electronic means of communication. 33he survey was mainly aimed at AEX and AMX-listed corporations.In 2006, none of the corporations offered a possibility for communication among shareholders or between shareholders and the Raad van Bestuur (board, hereinafter abbreviated as RvB) and the Raad van Commissarissen (board of supervision, hereinafter abbreviated as RvC) by means of a web board.
No post-testing has yet been made to see whether this has changed after the Wec took effect.Only a minority of corporations (40%) webcasted the AVA through a website.Some 55% of corporations that did not webcast their AVAs were also not planning to do so in the future.After the Wec took effect, the average percentage of corporations that webcasted their AVAs dropped to 36%.However, there was a big increase in AEX and AMX-listed corporations that started webcasting their meetings (67% and 65% respectively).The majority of corporations and investors do expect that the use of ICT applications will increase in the future.A real answer to the question whether the Wec has led to an actual increase in the use of ICT application and whether this use can be deemed to be a contribution to shareholder involvement remains uncertain.In 2006 Abma observed some alternative problems in the field of distance voting.He was of the opinion that the time between the publication of the relevant documents and the deadline to cast a vote was too brief (compare Section 1.2). 34Abma finds it quite remarkable that that participation of foreign shareholders is still obstructed by the fact that meetings are conducted in Dutch.Only 15% of the average listed corporations use English as the lingua franca. 35This is slightly more for AEX-listed companies, namely 24%. 36Van der Krans states that there is a limit to the number of participants in an active video conference.He also sees very little benefit in electronic voting, since casting a vote without following the meeting proceedings would go against the principle of the AVA as a platform for debate. 37The possibility to electronically cast a vote prior to the meeting could also frustrate corporate governance since it could put the AVA debate out of play before it has even commenced. 38he concept of a completely virtual VGM does not satisfy Dutch company law.Articles 2:116/2:226 BW clearly state that the AVA is supposed to be held at a physical location in the Netherlands.However, this does not mean that it would be undesirable to explore the possibilities for a VGM.In fact, a proper investigation into the theoretical concept of the VGM could be anticipatory and could at least lead to a wider scope of alternative solutions to the problem of shareholder involvement.

A VGM blueprint according to academic literature
This subsection will mainly focus on the VGM blueprint by Van der Krans, since he has incorporated different VGM models into a cohesive whole. 39Another ground for this choice is the fact that Van der Krans' blueprint fits the scope of this article since it has been constructed with respect to the Dutch system of corporate law and in such a way that it would convince the Dutch legislature. 40n the proposed model by Van der Krans, there would no longer be a physical assembly of shareholders, the VGM would be completely held online.The VGM would therefore be a heavily secured website which the shareholder could access with the use of a login name and password.This heavily secured website could be seen as a platform where the shareholder could not only study the agenda, the annual reports and other documents, but it would also be used to debate and to cast a vote. 41It is important to note that once the shareholder has logged onto the VGM website, he is not just 'virtually' present, he is to be considered as being legally present.Because of this virtual legal presence the shareholder's vote will have legal effect and thus be included for the necessary quorum. 42Presentations by the RvB/RvC will be podcast as video or Power-Point presentations.One of the most essential features of the VGM website would be the web board (also called a forum or bulletin board) on which shareholders could communicate amongst each other and the RvB.According to Van der Krans, the web board would be a good means to separately discuss different issues on the agenda and to introduce new points to the agenda, which would solve the asymmetric information distribution. 43The shareholder could cast his vote in a virtual ballot box during annually set voting periods, ranging from some hours to some days.After the voting period has terminated the results will be published and the meeting will be closed. 44During the VGM, the RvB would be responsible for the same tasks as at the physical AVA, namely to present its policy over the past financial year, to be held accountable for its performance and to answer questions.
Apart from authorised employees and members of the RvB/RvC, the VGM website should be accessible to three categories of users (each with unique user rights), according to Van der Krans. 45The first category consists out of 'Observers', who are strictly allowed to follow the proceedings and to look at the meeting papers.These 'Observers' would primarily be journalists, financial analysts, but could also be other parties with a special interest. 46The second category consists of 'Debaters'.These users can participate in the meeting but have no right to vote.Specifically persons who hold depositary receipts for shares will fall under this definition.In the third and last category one can find the 'Voters', who unequivocally represent the shareholders and other persons with voting rights. 47nother important aspect of the VGM is to maintain order and to formally control procedural rules. 48This task is currently reserved for the chairman of the meeting (usually a member of the RvC).For the VGM it would be of quite some significance if the moderation would be done by an external and technically qualified moderator.This special function should be executed objectively in order to safeguard shareholder's rights.Therefore a special position, called the Shareholder Rights Manager (SRM), is proposed.The SRM would be responsible for the technical, legal and organisational processes of the VGM.In order to avoid a conflict of interest and to maintain independence from the board, the SRM would be appointed by the meeting of shareholders.According to Zetzsche, this new allocation of duties between the SRM and the RvC can only lead to benefits.The SRM can facilitate the shareholders, while the RvC can focus more on its supervisory and advisory tasks.This could lead to an increased specialisation of the tasks in question. 49etzsche proposes a slightly different VGM which is held every quarter.A quarterly VGM could actually be tantamount to so-called institutional investor meetings which revolve around shareholder-to-management communication control.A quarterly VGM would be attended virtually.The board would broadcast from a studio and would engage in live debates with virtually present shareholders by means of video conferencing.Each meeting would end with a voting period.Furthermore, the VGM would consist of a permanent web board which would be moderated by the SRM.Additionally, the web board would contain summaries of the quarterly meetings and have an extensive Question and Answer catalogue. 50It would be advisable first to see whether it would be feasible to demand such an active attitude from the individual shareholder.Cox and Hazen have witnessed rapid developments in the US concerning the digitalisation of the AGM, with Delaware leading the way. 51Delaware was the first state to adopt a form of VGM in the Delaware General Corporation Code.Even though Delaware is a small state, 52 it is considered to be a corporate haven.Some 800,000 business entities have their legal seat in Delaware including more than 50% of all U.S. publicly-traded companies and 60% of the Fortune 500. 53he advantage of Delaware derives mainly from its size.Since the state's population is small, the per capita impact is large, allowing the corporate bar to exert a strong influence on the legislature.The input of Delaware citizens and pressure groups like unions are weaker than in other states since most of them are unaffected (because the corporations mostly operate elsewhere). 54This corporate climate has resulted in a very progressive corporation law, low state taxes and a highly specialized judiciary (Chancery Court of Delaware). 55n 1996, the Delaware-based corporation Bell & Howell started to webcast its Annual General Meeting online.The 950 shareholders that followed the proceedings of the AGM online were not actually legally present, and therefore did not attend the meeting for quorum or voting purposes. 56At this time, the law stated: '[meetings] may be held at such place, either within or without this State, as may be designated by or in the manner provided in the bylaws or, if not so designated, at the registered office of the corporation in this State.' 57The Internet did not fall under the traditional definition of a (physical) 'place' as the Delaware legislature had intended.

The use of electronic means of communication during the AGM
In reaction to these technological developments, the Delaware Code was amended in 2001 so that Delaware corporations could benefit from these technological developments. 58To the Del.Code Ann.Tit. 8 §211(a) was added: 'the board of directors is authorized to determine that the meeting shall not be held at any place, but may instead be held solely by means of remote communication (...)'.§211(a)(2) ensures that the VGM and the AGM are equal by stating subsequently that a shareholder can participate in the meeting by means of remote communication and should be deemed present in person. 59he board has sole discretion to determine whether a meeting will be held by means of remote communication.However, the virtual shareholder meetings should meet some requirements.The corporation needs to implement reasonable measures to verify whether persons participating in the meeting are actually authorised to do so. 60Corporations also need to meet the requirement that participants have a reasonable opportunity to participate (including the opportunity to read or hear in real time) and to vote. 61Apart from this the board has to keep records of shareholder actions to prevent shareholder fraud. 62The shareholder needs to confirm his identity and the authenticity of his share ownership.However, the DelGCL does not provide any additional (minimum) procedural rules on the manner in which this should be done.
Pursuant to Del. Code Ann.Tit. 8 §232, it is possible to give notice of a planned shareholder meeting by electronic transmission.The law does require that the electronic transmission is consented to by the stockholder and that the notice is sent to a number or address at which the stockholder has consented to receive notice (Del.Code Ann.Tit. 8 §232(b)).It is also possible to publish notice of a planned shareholder meeting on the corporate website, as long as a separate notice is sent to the stockholder of such specific posting.The DelGCL does not give stockholders the right to access a list of email addresses of other stockholders. 63Due to this, little internal communication among stockholders is possible.

How effectively does the VGM function in Delaware?
When attempting to answer the question how the legal VGM was put into practice in Delaware, one is hopelessly limited by a shortage of concrete cases.Consequently, no serious quantitative research has been conducted on the possible contribution of the Delaware VGM to corporate governance.
In 1996, Bell & Howell started to webcast its shareholder meetings because the corporation wanted to boost its image as a high-tech company. 64According to Bell & Howell the relatively low costs (between $ 10,000 and $ 15,000) makes the VGM an attractive option.A factual economic evaluation of the Delaware VGM has not yet been conducted. 65The attendance rate at AGMs, as measured in votes, has been quite high in the US over the past decade, namely around 85%. 66 However, the actual attendance rates, as measured in actual participants, remain far lower than the number of votes cast, namely 25% of the total number of stockholders. 67The actual use of electronic means of communication in 2003 does also not give much reason for joy.Less than 50% of US corporations publish the presentation of the board at the shareholder meeting online. 68here are only some qualitative examples of the VGM in practice.Inforte Corporation held its first entirely virtual shareholder meeting in April 2001. 69The participants could ask questions to the board via email. 70The board of Inforte answered all the questions and published several PowerPoint presentations on its VGM website. 71The VGM was pronounced a success and the board decided to continue to hold online shareholder meetings in the future.No reports are available on Inforte's subsequent shareholder meetings. 72Cyber Inc. followed in Inforte's footsteps in 2002, hoping to give their attendance rate a boost (from less than 10 out of 28,000 shareholders!). 73Both Cyber Inc. and Inforte did not make a web board available for shareholders and the board to debate. 74Boros states that the unpopularity of the VGM in Delaware is caused by the fact that the DelGCL does not set concrete procedural requirements for a virtual shareholder meeting. 75Delaware law does not require a web forum to provide an electronic analogy to the social confrontation of a physical meeting.Van der Krans lists three other possible causes for the apparent low attractiveness of the VGM. 76First, there is no supervision of the posting and removal of electronic contributions by shareholders.Secondly, the board offers no guarantees that questions will indeed be answered.Thirdly, the board has the discretionary power to decide to hold a VGM.According to Van der Krans this could lead to scepticism among shareholders since they have not been involved in the decision-making process. 77he VGM also received quite some negative publicity in the American media.This ultimately led to the withdrawal of a proposed bill to hold online meetings in Massachusetts.
Another possible reason to explain the fact that the VGM enjoys poor popularity is the fact that corporations tried to protect themselves against the so-called written consent option (Del.Code Ann.Tit. 8 §228).This procedural right does not directly relate to the use of electronic means of communication, but ensures the possibility that shareholders could reach decisions without a meeting in certain special situations.Since this written consent option sometimes leads to shareholder hostility, corporations tend to deviate from this legal provision by making physical shareholder meetings obligatory. 78s Van der Krans acknowledges, the aforementioned explanations constitute only a logical examination of the possible causes for the supposed ineffective functioning of the VGM in Delaware. 79Much further empirical research is needed to conduct a proper evaluation.This could not only be meaningful from a European/Dutch point of view, but it could also be a valuable reflection for Delaware itself and other states in the US that would be susceptible to a VGM.From a comparative legal perspective, it could contribute fruitful results which would give us the relevant legal understanding. 80

The use of electronic means of communication in Denmark
In 2002 the Danish Government launched a modernisation of its company law under the title of 'freedom with responsibility'.This modernisation scheme was part of its overall action plan to simplify administrative facilities. 81The modernisation resulted in the law on limited liability companies and the law on private limited companies being amended (LOV no.303 af 30-04-2003).The law amendment was intended to improve active ownership (aktionaerdeltagelsen).The view of the Danish legislature was that the most central forum for shareholders' active ownership is the general meeting. 82Another goal was to clarify the law on the use of modern information technology for corporate-shareholder communication. 83For a while there was uncertainty in the Danish legal literature about the question of whether or not §65 of the Companies section required a physical meeting and if it was allowed to use electronic forms of communication. 84Among other issues, the law included the following elements with regard to information technology: 1.
It introduced the possibility for shareholders to participate and vote in a delvis elektronisk Generalforsamlingen (partial electronic General Meeting), without the requirement of a physical presence (LOV no.303, §65a stk.1).Unless the articles of association provide otherwise, the Bestyrelsen (board) may decide that shareholders can be virtually present at the meeting.In a delvis elektronisk Generalforsamlingen, the physical meeting is still required and electronic access functions as an addition to a physical appearance.
The General Meeting may decide on a fuldstaendig elektronisk Generalforsamlingen.This resolution is subject to some stringent criteria.It should be upheld by at least two-thirds of the votes cast, and a quorum of two-thirds of the share capital applies (LOV no.303, §78).
It is subject to the condition that shareholders representing 25 per cent of the total voting share capital should not vote against the resolution, and the articles of association may submit the resolution to additional requirements (LOV no.303, §78 stk.2-3).Furthermore, the resolution must contain information on the way in which electronic systems are used (LOV no.303, §65a stk.2) and must be included in the articles of association.

3.
Shareholders who participate electronically in a partial or complete General Meeting have the right to ask questions on the agenda or on documents before the expiry of a deadline prior to the Generalforsamlingen, as laid down in the articles of association (LOV no.303, §65a stk.5).The same stringent measures from LOV no.303 §78 stk.1-3 apply when amending the deadlines.It is for the companies to create a detailed framework for the procedure and conduct of the electronic meeting. 85The Bestyrelsen can determine the requirements for the electronic systems used for the partial and complete electronic General Meeting (LOV no.303, §65a stk.3).The system must be designed in such a way that the General Meeting is safe and that it meets statutory requirements (LOV no.303, §65a stk.4).Another requirement of stk. 4, which the electronic system should meet, is the fact that it must be reliable in determining the shareholders who participate, the capital, the voting rights which are represented and the voting results.It is of paramount importance that the shareholder has the full possibility to participate, vote and express himself.He is aided by the requirement that the corporation must give sufficient notice to convene the meeting on how to sign up and provide information on the procedure in connection with electronic participation (LOV no.303, §65a stk.3).

4.
The Generalforsamlingen may decide that the company and shareholders may use electronic documents and electronic mail in their communications (electronic communication, LOV no.303 §65b).The resolution must be clear which messages are covered by the decision and the manner in which electronic communication must be used (LOV no.303 §65b stk.2).However, where the law prescribes that communications have to be made by a public call or gazette, electronic communication does not replace the official invitation or notice (LOV no.303 §65b stk.5).

5.
The Generalforsamlingen can, furthermore, decide that shareholders must be available, including by means of an electronic medium (such as a web board, LOV no.303, §26 stk.3).This would then make it easier for shareholders to communicate with each other in order to organize their votes and form coalitions.86

How effectively does the electronic General Meeting function in Denmark?
It is still too early to conclude whether partial and complete electronic meetings have functioned effectively in Denmark.After the amendment of the law, corporations first started to amend their articles of association in order to vote on a possible partial or complete electronic meeting.Exact figures on the number of amendments to articles of association and resolutions on partial and complete electronic meetings are still required in order to formulate meaningful conclusions.
In the absence of real scientific data on the effects of the law amendment, one can only rely on messages provided by the Danish media.No mention has been made of the first ever Danish complete electronic General Meeting.Sparindex, however, is reported to be the first Danish corporation to plan a purely electronic extraordinary meeting in August or September 2008. 87parindex hopes with this initiative to boost its attendance rates to one-third of the total voting capital.It admits that the complete electronic meeting does not save any costs.However, according to Sparindex the aim of the electronic meeting is to improve the dialogue with its shareholders by providing reasonably quick answers.It must be reported, however, that Sparindex will physically retain its General Meeting and is planning to decide on a partial electronic General Meeting.
There are also no reports on the holding of partial electronic General Meetings.This could nevertheless be explained by the fact that a partial electronic meeting does not entail any major differences from the conventional meeting, since a physical meeting is still required.However, the prospects of the proposal for the electronic meeting were to the subject of serious investigation in the spring of 2002.The Danish Public Administration contacted several groups and major public limited corporations to assess the interest in an electronic General Meeting.The research confirmed that a large majority were in favour of a voluntary alternative to the physical General Meeting. 88It could be plausible to assume that these corporations had amended their articles of association to enable a resolution in favour of an electronic General Meeting to be adopted. 89

Some comparative conclusions and explanations
When comparing the Delaware Virtual General Meeting with the Danish elektronisk Generalforsamlingen some striking differences emerge.Denmark differentiates between three models for holding a general meeting, the conventional General Meeting, the partial electronic General Meeting and the complete electronic General Meeting.Delaware mainly differentiates between two models, a physical meeting and an entirely Virtual General Meeting.While in Delaware it is the board that decides whether to hold an entirely Virtual General Meeting, in Denmark the Generalforsamlingen is the competent authority.The Danish Bestyrelsen may only decide on a partial electronic General Meeting.Another obvious difference is that the law of Delaware provides very marginal requirements for the decision to hold a VGM while the resolution by the Generalforsamlingen is submitted to qualified majority voting and a quorum.
Concerning the procedure, one could note some similarities.Both the laws of Delaware and Denmark require that the shareholder can participate and vote in the meeting.However, the nature of 'participation' is different in both legal systems.In Delaware this means the opportunity to read or hear, while in Denmark the shareholder should also be able to express himself and ask questions.Danish law is also more specific in the duty of the Bestyrelsen to give shareholders access to the electronic meeting (the system must be safe, reliable and the shareholder should get enough information) than the law of Delaware, which gives the board the duty to give partici-

Introduction
The Dutch legislature has not yet been convinced of the use of a VGM in the Netherlands. 93owever, it is believed that the VGM will entail the following positive causal chain: the VGM will lead to higher attendance rates; this will lead to better corporate democracy, which will, in its turn, advance corporate governance ultimately leading to improved corporate performance.This positive causal chain is usually presented as an obvious result of the VGM.It should be noted, nevertheless, that the links of the causal chain could be questioned and much will depend on the concrete legal and practical realisation of this modern construction.Nolan, therefore, keenly stated that the legislature should (i) obtain a clear overview of the effect of shareholder engagement on corporate governance, and (ii) have a clear understanding of the present law and the practice which builds on that law. 94To elaborate these two basic thoughts, this section will assess the existing hypotheses with regard to the potential contribution of the VGM to corporate governance and propose means to empirically measure these hypotheses.

Causal presumption 1: the VGM will lead to higher attendance rates
The desirability of the VGM is typically grounded by the causal presumption that the VGM will lead to higher attendance rates.The length of the proceedings, the absence of travelling time and costs, and an improvement in the information distribution would mean that more shareholders will participate.However, this presumption could be questioned due to two reasons.
First, although the possible causes of the low involvement of shareholders (transnational voting, rational apathy, asymmetrical information and the unfortunate planning of the AVAs), as brought forward in the academic literature (Section 1.3), could be considered quite apparent, it is not clear what criteria and methods were used to establish a causal relationship between these causes and the low involvement.This also leaves doubts concerning any possible other causes that could have contributed to low involvement.As the aforementioned proposed causes could constitute a logical ground for a VGM, it would be wise to be of the utmost certainty concerning their existence and effect. 95econdly, it is unsure whether a positive causal effect exists between the VGM and the attendance rates.The opposite effect could be illustrated by a qualitative study, undertaken by Unerman and Bennett, of Shell's failed attempt to improve Internet stakeholder dialogue through the use of a web-forum. 96This failure was caused by the mere fact that almost no stakeholder was willing to actively participate and contribute to the web-forum.The main question is therefore with regard to the VGM whether a shareholder will feel committed to contribute to a virtual shareholder meeting.One could say that a web board will only lead to further disengagement since the decisions are not the 'fruit of face-to-face consultancy'. 97Catasús and Johed define the annual shareholders' meeting as a social ritual that re-emerges every year.During this social practice the past accounting year is ritually closed and different voices are heard. 98Therefore, one needs to be aware that the abandonment of this tradition with the introduction of a VGM could have a possible negative effect on the attendance rates of the VGM.One should not underestimate the power of face-to-face gathering as a motivation for participation in the shareholders' meeting.This statement might seem somewhat paradoxical, since in 2007 only 45% of the capital was represented during the AVA.However, with regard to the disappointing experiences with the Shell web-forum, one could state that due to the power of face-to-face gathering at least 45% of the shareholders did turn up.
There are many other possible factors that could negatively influence Internet usage and thus the attendance rates at the VGM.The first and obvious one could be ignorance with regard to the use of IT applications.Also age seems to have a significant negative effect on Internet usage. 99hirdly, the presumption of a positive causal relation between the VGM and higher attendance rates can be questioned if alternative hypotheses, variables and explanations are not examined.This could increase the danger of internally invalid inferences.A good example of alternative variables that have a positive influence on attendance rates was already mentioned in Section 1.2.Eumedion claimed that an increase in attendance rates between 2006 and 2007 was explained by the fact that companies published their agendas earlier before the meeting and the fact that the registration terms were brought forward. 100Although these causes have not yet been confirmed by other studies, it would be wise to include a broader spectrum of alternative variables that could lead to an increase in attendance rates.

Causal presumption 2: VGM and high attendance rates will improve corporate democracy
Van der Krans and others 101 state that the VGM and the resulting high attendance rates will improve corporate democracy. 102Since more shareholders will attend the VGM, they will be most likely to form alliances.The VGM seems to be an effective means for minority shareholders to unite and pose a counterbalance against institutional and majority shareholders. 103Additionally, the VGM would solve the asymmetrical information problem, leading to higher corporate democracy.This second causal presumption again raises some questions.
First, one could question the value of the outcome variable, corporate democracy.The concept of corporate democracy has been widely criticised.Epstein considered corporate democracy as a common myth since the shareholder has no alternatives to choose from.Another objection against corporate democracy is that large corporations function too inefficiently for shareholders to reform.For corporate democracy to work as a measurable outcome variable it would be sensible to create a workable definition.
Secondly, it is questionable whether attendance and participation can be considered as the imperative factor to improve corporate democracy.Increased shareholder participation constitutes only a small part of the solution to the corporate democracy problem.Much of the problems surrounding corporate democracy also lie in the number of rights and alternatives that the shareholder does (not) have.Irvine and Epstein state that corporate democracy is a myth since the shareholder has no alternatives to choose from. 104Coffee states that corporate democracy does not work because large multinationals are often too inefficient to be reformed. 105Not only attendance rates but also procedure can contribute to corporate democracy.In the case of a virtual shareholder meeting this would be largely dependent on the practical and concrete realisation of the VGM by the company.It would be plausible to assume that if corporations provide marginal possibilities for shareholders to find each other and form coalitions this would have adverse effects on corporate governance and attendance rates.
Thirdly, it would be wise to consider studies on Internet usage, Internet involvement and web forum usage.It would be a doubtful practice to put absolute faith in the proactive involvement of the online shareholder to improve corporate democracy.Attendance does not necessarily guarantee a high degree of involvement with regard to debating and coalition forming.The level of proactive involvement could depend on many variables such as age, gender, perceived convenience, perceived usefulness, technical difficulties et cetera. 106Psychological research has also shown that Internet use adversely affects social involvement and psychological well-being. 107t would be wise to investigate if this does not pose a serious obstacle to active contribution to the web board.

Causal presumption 3: higher corporate democracy will lead to more corporate governance
The third causal presumption is that improved corporate democracy will have a positive effect on corporate governance.This is based on the argument that since shareholders are more easily able to exercise their rights, the AVA can claim its righteous position in the corporation.This would then lead to a better system of checks and balances, and thus corporate governance. 108Also this third causal presumption raises certain questions.
First, as with corporate democracy, it is difficult to measure 'good' or 'better' corporate governance without clear concepts. 109It could be equally or even more intricate to measure 'improved' corporate governance as a dependent variable of the governance attribute 'corporate democracy'.It is of some importance to note that quantitative studies on 'good' governance usually do not consider attendance rates and corporate democracy as governance attributes. 110his does not mean that it is wrong to view these variables as governance attributes, but it does stress the need for careful deliberation.
Secondly, corporate democracy quite often fails in practice because the board does not always follow the needs of the shareholders.Therefore, more corporate democracy does not have to lead to more corporate governance.A possible legal regulation of the VGM procedure should therefore include the lessons learnt on the corporate governance theories.If the VGM proceedings would be submitted to mandatory provisions, the board would be forced to act more as a steward in the service of the corporation and its shareholders.
Thirdly, in line with the first question against the third causal presumption, it would be indispensable to take into account the tremendous reservoir of other independent variables that could have an effect on corporate governance.A discussion on the VGM and its supposed (in)direct positive causal effect on Corporate Governance cannot take place without considering corporate governance attributes as a whole.The supposed boost of the VGM to corporate democracy is not likely to contribute to corporate governance if, for example, stock-incentive plans are adopted without shareholder approval.When evaluating the VGM in the light of other corporate governance attributes it advisable to distinguish between attributes that could (i) due to their (non-)existence obstruct the working of corporate democracy as a contributor to corporate governance, and (ii) be more suitable, economic or functional contributions to corporate governance compared to the VGM corporate democracy chain.The VGM could, for example, be combined with a progressive executive remuneration system.

Causal presumption 4: more corporate governance will lead to higher corporate performance
The fourth causal presumption is that more corporate governance will lead to a higher corporate performance.Van der Krans bases this causal presumption on Diacon's and O'Sullivan's statement that certain corporate governance attributes can lead to better financial performance. 111owever, again some strong questions can be posed since this causal presumption could create, when taken without due consideration, false expectations.
First, in this final causal presumption, some confusion arises as where to place the different variables: the VGM, attendance rates, corporate democracy and corporate governance in relation to corporate performance.Would it be the aim to directly assess the effect of the VGM on corporate performance?Or is corporate democracy regarded as a corporate governance attribute which affects corporate performance actually assessed? 112The causal presumption suggests, actually, that corporate governance as a whole could be considered as a corporate performance attribute.However, Diacon and O'Sullivan took several corporate governance attributes and measured their effect on corporate profitability.It is therefore not 'corporate governance as a whole' but (a combination of) corporate attributes that could have some impact on profitability.
Secondly, the study by Diacon and O'Sullivan could be considered to base conclusions on the contribution of the VGM to corporate performance.Their study categorized types of corporate governance attributes that were in use in the U.K. insurance industry.Diacon and O'Sullivan concluded that some governance factors had a positive effect on performance measures.Corporate performance was affected by the number of directors, the system of remuneration, and the existence of audit and mutual status. 113Also the appointment of the right chairman and a high proportion of ex-executives could have a sufficient effect on the performance of a corporation.However, it is important that, in general, the governance factors had a non-linear impact on corporate performance factors.Formal governance instruments could even inhibit profitability to extreme levels. 114The effect of different factors on performance is believed to be complex, non-linear and dependent on the individual business.It would be sensible to impute only a modest role to the VGM in this process.

Lessons to be learnt from Delaware and Denmark
Even though the real contribution of the VGM to corporate governance has not yet been convincingly proven, it is based on a plausible logic to some extent.Based on the emphasis on selfregulation in Dutch corporate governance, it would be wise to involve the shareholder as much as possible in order to create a good system of checks and balances between agent and principal.If the VGM would contribute to the involvement of the shareholder it should be seriously considered by the Dutch legislature.Much of the potential of the VGM seems to be dependent on the way in which the law regulates the VGM and the commitment of the shareholder, as well as other controlling variables.It would therefore be useful for the Dutch legislature to take into account some lessons learnt from Delaware and Denmark. 115The experiences with the VGM in Delaware seem to have been anything but promising, and even though Danish law seems to give some more valuable models for a VGM, its function in practice remains unclear.
The first problem was that the DelGCL leaves many procedural issues unarranged.As a result the corporation has a wide degree of freedom to organise the proceedings of the VGM so as to be detrimental to the needs and rights of the shareholders.In this respect we can point to the fact that the DelGCL does not contain any mandatory provisions for the corporation to include a web board to facilitate the debate between the board and shareholders. 116With a view to the protection of corporate governance relations it would be wise for the Dutch legislature to emulate the example of Danish law and regulate procedural aspects of the VGM as inclusively as possible.Dutch law should probably be of a mandatory nature and include the requirement of a web board.Secondly, the law needs to be clear.The DelGCL unfortunately seems to leave too much room for interpretation by using terms such as reasonable opportunity.This term is most likely to be submitted to a restrictive interpretation by the corporation, to the detriment of the rights and needs of the shareholders. 117The third problem is that the AGM in Delaware has no say in the decision-making process to hold a VGM.However, it would be prudent to submit a proposal to hold a VGM (including its proceedings) to the shareholders' meeting for approval.Otherwise the VGM is nothing less than a means to create a bigger distance between the board and the shareholders.However, the VGM revolves around the principle of shareholder involvement.When this is ignored, by not involving the shareholders in the decision-making process, the VGM is stripped of its ideological function and necessity.Consequently, the danger arises that the VGM is used by the board as an extra barrier for effective shareholder control.It would be wise to consider, therefore, a restrictive option to hold an entirely online VGM, such as in Denmark.The result might then be that, as in Denmark, it will be quite unlikely that a corporation will decide on a completely online VGM.Fourthly, the legal system in which the VGM is implemented needs to have an attractive framework for corporations to hold a VGM.The law needs to be free from internal contradictions and ambiguities that could pose an obstacle to the functional application of the VGM.Internal contradictions in the DelGCL, such as the anomaly between the written consent and the VGM (Section 3.2), that can lead to shareholder activism or shareholder hostility could mean that the VGM is not an attractive option for corporations.Fifth and finally, what might be lacking in Danish law is an escape option, which the General Meeting could use when the completely online VGM fails in its attempts to attract shareholders.Decisions to change back the articles of association in the case of failure, are also subject to the same stringent measures in LOV no.303 §78 stk.1-2.Theoretically, in case of a complete failure (such as the Shell web board, Section 4.2.1) of the VGM a deadlock situation could exist in which the shareholders' meeting cannot amend the articles of association because it is not able to meet the necessary requirements due to the failure of the VGM.Dutch law should therefore foresee an escape possibility.

Economic feasibility
It might be of some importance to reflect on the possible economic attainability and consequences of the VGM.Academic literature has frequently addressed the possible costs related to the VGM. 118It seemed that the general assumption is that in order to hold a VGM, the corporation needs to make substantial investments in technical and organisational facilities.This view probably disregards healthy market forces.Like in the US (Shareholder.com),the market will probably foresee complete ICT provisions if the VGM would be allowed by law.If the corporation would indeed choose to host the VGM on its own servers, it could rent additional computer power to be able to handle the increased data traffic during the VGM peak. 119The SRM, to be appointed by the AVA, already exists.This position is currently covered by the PR or Investment Relations manager (or department), responsible for the organisation of the physical shareholders' meeting.
The (foreign) investor is believed to profit the most from the VGM, with regard to travel and housing costs as well as lost income. 120It is believed that shareholders do not need to have their costs reimbursed since they are deemed to dispose of the necessary media to attend the VGM.
Possible benefits of IT developments are usually unclear or doubtful.IT systems could limit the flexibility of the VGM (IT is the actual condition for a functioning VGM).IT developments are driven by technology and not by law.Also inefficient relations between IT users and IT developers, as well as long-running times of implementation projects could limit the economic advantages of the VGM.Sometimes the running time of the implementation project is longer than the running time of the technology itself: these issues require IT governance.121

Preliminary conclusions
The effects of the VGM blueprint on shareholder involvement and corporate governance are difficult to predict.The potential of the VGM to contribute to corporate governance seems to be dependent on the commitment of the shareholders and the possibilities that the law provides.Regarding the former, too little research has been conducted into the question whether the implementation of the VGM in Delaware and Denmark has been successful (Section 3.5).With respect to the latter, however, one could learn some valuable lessons from Delaware law.It is essential to note that certain specific VGM models could even contribute to worse corporate governance if it practically leads to the total abolition of the shareholders' meeting (Section 3.2).This potential adverse effect of the VGM should be submitted to a serious investigation.
The extent of the contribution of a VGM to corporate governance and ultimately corporate performance remains questionable.There are numerous alternative dependent and independent variables that could influence corporate governance at various levels.The VGM could probably at most contribute partly to a better system of corporate governance.The causal chain as was proposed in the academic literature, and which was examined in Section 4.2, does not seem to provide convincing proof of the positive effects of the VGM on corporate governance and ultimately corporate performance.Alternative measures should be formulated to supply the Dutch legislature with convincing evidence of the benefits of a certain VGM model.It would be sensible to impute a modest role to the VGM, as a corporate governance instrument to improve corporate performance (Section 4.2.4).Corporate governance is not only dependent on the participation of foreign investors and audit mechanisms, also the number of board members, the system of remuneration, external audit and subsidiary ownership have their effect on the performance of a corporation.

Measurement proposal
The most suitable institution to supervise an empirical investigation into the possibilities of a VGM would probably be the Monitoring Commissie Corporate Governance (the Frijns Committee).Given the number of associated research and evaluation centres that are annually commissioned to conduct evaluative studies, the Netherlands probably has the necessary scientific instruments to undertake such an investigation.When proposing measurements it would be wise to differentiate between studies that would assess and predict the potential contribution of certain VGM models and the probable success rates, and studies that would evaluate the effects of an actual VGM on corporate governance after the amendment of the law.The emphasis in this proposal will be on the former.
To narrow down the scope of the studies it would be wise to eliminate some links of the causal chain, as was described in Section 4.2.This would decrease the possibility of making impossible inferences and having to take into account an exuberance of alternative variables.If one is to measure the effects of the VGM on corporate democracy or corporate performance these should each be part of an individual study and not part of a long and impractical causal chain that revolves around corporate democracy.This would probably improve the quality of the studies on the VGM if the notion of corporate democracy would be removed from the causal chain since it is seriously disputed in the academic literature and it would be difficult to measure as an attribute of corporate governance.Furthermore, to formulate empirical predictions on the (indirect) effect of the VGM on corporate performance would be another unworkable exercise and would impede the general objective of the VGM, namely to improve corporate governance.Diacon and O'Sullivan even stated in their conclusion that: 'Although we have isolated the constituent factors in the system, their independent impact on company performance is complex, highly nonlinear, and dependent on the nature of the business transacted.It is evident that there can be no universal recipes for the best form of corporate governance, even if the focus of attention is confined to just one industry.' 122tudies would be more compelling in the prediction stage preceding the law amendment if they would assess different VGM models and their diverging/converging effects on attendance rates and corporate governance.This would give the legislature a more complete overview of workable and unworkable models in order to come up with an optimal VGM model.Most of all, it would show which variables of a VGM model would have a negative impact on attendance rates and corporate governance.Moreover, it would be wise to differentiate between potential causal/correlational links to be investigated.For example, one could examine the expected attendance rates of various VGM models, or one could focus on the effect of higher attendance rates on corporate governance, or the assessment of the VGM as a direct corporate governance attribute.

Predicting the effects of the VGM (before amending the law)
To gain more insight into the low attendance rates it would be sensible to investigate the more plausible causes of low involvement and to verify the causes that have been mentioned in academic writing (Section 1.3).This would create a common frame of reference in which possible solutions could be proposed (of which the VGM would be one option).As very little research has been conducted in Delaware and Denmark on the effect of the VGM law, one lacks different data on which valid and generalisable predictions can be based.It could be effective to carry out a detailed survey among shareholders and board members to assess their intended engagement in various VGM models, as was done in Denmark (Section 3.4).In this way one could come to more meaningful conclusions on shareholder engagement motivations and intentions.
Another study could aim to examine the possible relation between attendance rates and corporate governance.This would not include the measurement of VGM models but would, however, provide a clear overview of alternative variables that lead to the increased attendance rates that have been observed over the past few years.This could be very clarifying in the light of previous studies on good corporate governance that did not include attendance as an attribute.If corporate governance is certainly positively affected by higher attendance rates it could be a strong argument in favour of the VGM if it would indeed increase attendance.This study alone could already be pioneering since it would shed new light on corporate governance.

Evaluating the effects of a VGM (after amending the law)
In the evaluation stage of a VGM, the law has been amended and VGMs have been held.At this stage one has more certainty, as the effects and consequences have come apparent.The research design that could be adopted is a natural post-test experiment.The VGM is in this case considered to be an event that has occurred naturally.A wide range of alternative variables and explanations could be taken into account when measuring the effects of the VGM in this phase.It would be wise to include a control group of corporations that have not adopted the VGM model.Data could be derived from surveys among shareholders and corporations as well as other statistical data, such as attendance rates et cetera.Closely related would be cross-sectional research in this respect.Case studies on specific corporations could be included so as to gain deeper knowledge of and to find sophisticated explanations for the causal and correlational effects of the VGM.It is at this stage that one could perhaps consider bolder but workable hypotheses assessing indirect causation, including well-defined variables such as corporate democracy and even corporate performance.Still, one should be careful with the meaning of variables in cross-sectional and post-test experimental designs regarding the VGM.The danger could arise that based on the mere fact that there is a correlation between the VGM and attendance rates, this would indicate a causal relation. 123

Experimental design
The ultimate theory-testing strategy could be realised through an experiment in which a VGM would be simulated under the current state of the Dutch law.The votes, casted by the online shareholders, would then not be included in the legally binding decision, as the Delaware corporation Bell & Howell did in 1996.In this way, a pre-test (measuring attendance rates and/or other corporate governance attributes) could be conducted in the preceding accounting year and a post-test after the intervention, namely the VGM.Again, a group of randomly picked corporations that have not held a VGM would function as a control group to prove or disprove alternative influencing variables or explanations.The problematic part of this proposal is of course the cost and the lack of any direct gain for the corporations involved.However, it could function as a good promotion stunt for a Dutch technology firm comparable to Bell & Howell. 12425 This subject could also be subsequent to further investigations.It is a fundamental issue which arises in every debate on the integration of modern means of communication in positive law (examples: electronic signature, distance selling, e-commerce, electronic voting).The more virtual and extensive the technology, the more sovereign its nature (examples: online gambling, cybercrime, spam et cetera).If states do not want their sovereignty to be reduced by a virtual Wild West, they will need to find some sort of means to regulate it.126 The expectation is that this trend will continue in the near future (Section 2.2.2).An increasing number of Dutch corporations already host a VGM website where shareholders can access documents and webcasts with regard to the AVA.127 Supra Section 4.2.2.128 Zetzsche, supra note 39, p. 57.129 Van der Krans, supra note 2, pp.280-281.185

General framework
In Section 4.6 it was suggested that empirical studies would be more compelling if they would assess different VGM models and their diverging/converging effects on attendance rates and corporate governance.Only evaluating a possible VGM based on one blueprint would be too limited, given the number of possible alternative positive variables on corporate governance, as well as possible adverse effects of certain VGM models on corporate governance.Therefore, to serve the continuing discussion on a VGM, this section will propose two models for the VGM that represent two ends of a spectrum of virtuality and law.Between these two ends, there could be a broad range of modalities in which corporations use electronic means of communication to organise their shareholders' meetings.The law would play a fundamental role in this spectrum and would be parallel to virtuality (more law in proportion to more virtuality).The first and lower end of the spectrum (model 1) supposes a physical shareholders' meeting as is currently common, with the integration of modern IT applications.This form of IT backing seems to require little legal regulation and could be left to market forces.The (somewhat extreme) higher end (model 2) supposes a virtual world (such as Second Life) in which the VGM is held.The law is supposed to play a more active role since the 'non-existing' virtual world has actual legal effects in the offline world and should, therefore, be bound by worldly norms. 125

Model 1: integration of IT in the physical AVA
As stated before, AVAs increasingly rely on electronic means of communication during shareholders' meetings. 126The least progressive model would entail a continuation of the current practice along its main lines.However, some changes should be made to improve and to stimulate the proceedings of the VGM.During the meetings English should be used as the lingua franca (Section 2.2.2).Subsequently, a secured VGM site should be launched, including a web board, webcast channels and a virtual closing section.Ideally, the only means through which all shareholders could equally contact the board and vice versa would be by using the web board.In this way, the power of the institutional shareholders preceding the AVA could be further restricted, thereby providing room for more democratic deliberation processes. 127During the VGM, it would be possible to attend the physical meeting interactively by means of video conferences and online voting facilities.The actual meeting would be presided over by a member of the RvC or RvB, but with the support by a SRM (department).According to Zetzsche the Shareholder Rights Manager (SRM) needs to be appointed by the shareholders' meeting, so as to avoid any conflict of interest with the board. 128The new allocation of responsibilities between the SRM and the RvC can only lead to more advantages for both parties. 129The SRM could focus on the facilitation of shareholders, while the RvC could fully attend to its supervising and advising duties.The SRM would be responsible for the management of the VGM website and would link the physical and virtual contributions to the meeting.He could function as some sort of show host.The SRM would be the representative of the virtually present shareholders.At certain moments the meeting would be taken over by the SRM and he would phrase the most important views of the virtual attendants.This could be executed by making a critical selection of contributions from the web forum or by presenting poll results expressing the views of a certain proportion of the virtually present capital.The performance of the SRM would have to serve the shareholders since he is appointed and monitored by the AVA.The 'virtual shareholder' could be 'present' through various large screens in the conference centre, whether it be by image, webcam or probably merely as a 'statistic'. 130This model is already be quite workable since it would not require any amendments Articles 2:116/2:226 BW.

Model 2: towards a Second Life VGM
Model 2 supposes another (higher) end in virtuality: the 'Second Life' VGM in which the shareholders' meeting is held in a virtual world. 131Virtual worlds such as Second Life have been discovered by corporations from 2005 onwards as 'places of trade', and deserve to be taken seriously by lawyers. 132One should suppose a relative or abstract 'Second Life' model.In other words, the name 'Second Life' is chosen because it would appeal to the reader's imagination.It could be replaced by any other random denomination of a virtual world.The question remains whether a shareholders' meeting in a so-called Second Life environment would be realistic. 133t this point in time one can observe two developments in virtual worlds, (i) the settlement of commercial enterprises that market products and services, and (ii) the demand for legal solutions. 134Even though these two developments show that virtual worlds are no longer simply foolish computer games, the time of the Second Life VGM is not yet here.At this moment the developments are still too premature to hold a serious VGM in a virtual world.The primary reasons against this would be insufficient security and unfamiliarity. 135Especially in the field of security and procedural rules, the law would have to provide regulations to ensure accurate proceedings and the protection of company-sensitive information.The reason for this would be that up to now virtual worlds such as Second Life are not well protected against all types of intrusion.The main advantage of this second VGM model could be that the shareholder would feel more involved with proceedings because he would experience the meeting spatially (as opposed to a static VGM website without any live proceedings). 136This spatial experience could contribute to the feeling of 'social gathering', which could be an incentive for the greater involvement of the shareholders.The shareholder could, as with a static VGM, consult a VGM website at any time to receive all the necessary information in order to follow the proceedings.

Conclusion
The AVA plays an essential role in the supervision of the RvB and the functioning of the Dutch corporate governance system which is based on self-regulation.It is therefore of importance that shareholders turn up during the annual AVA.However, studies conducted in 2007 showed that on average only 45% of shareholders cast their vote during AVAs.Information Technology could increase the participation of shareholders.The possibilities that the Dutch Electronic Means of Communication Act offers are only marginally put into practice by corporations.Therefore, several academic writers have proposed a Virtual General Meeting (VGM), which would be completely held online.The shareholder could, consequently, access documents, debate and vote by means of a protected VGM website.At present, Dutch law does not offer the possibility of a VGM.In Delaware the experiences have been anything but promising.
Yet, the VGM is based on a certain degree of plausible logic.The Dutch legislature probably cannot escape the fact that it should allow more virtuality into the AVA in the near future.However, the possible positive and negative effects of the VGM have not been convincingly proven.More empirical research should be carried out in the field of attendance rates and corporate governance and the possible effect of the VGM on these variables.Different VGM models should be investigated, compared and evaluated.To find a suitable VGM model, the legislature could move in a spectrum of virtuality,137 with two (somewhat extreme) models at both ends: IT integration in the physical shareholders' meeting (minimum) and a Second Life VGM.Eventually, these developments are not to be shaped entirely by the lawyer.The demand for a VGM will necessarily result from market forces and the need to include foreign investors in corporate governance processes.The extent of this demand, as well as the possibilities to create a VGM that contributes to corporate governance, should first be submitted to serious quantitative research.