The Cayman Islands Grand Court has delivered its ruling in Re Shanda Games Limited (FSD 14 of 2016, 25 April 2017, Segal J).
This is only the second merger fair value appraisal ruling in the Cayman Islands, and in both cases Maples and Calder has represented the successful dissenters.
In Shanda, the Court found that the fair value of the shares owned by the dissenters (which were all funds managed by Hong Kong-based fund manager Maso Capital) was more than double the consideration offered in the take-private scheme by way of merger.
In summary, the Court:
- broadly endorsed the approach taken by Jones J in Re Integra;
- affirmed the principle that fair value in these cases should not incorporate a minority discount;
- demonstrated a willingness to investigate complex and technical valuation issues to determine the inputs in the experts' discounted cash-flow models, including the beta; and
- drew certain negative inferences against the company based on its failure to provide adequate disclosure or answers to the dissenters' expert's questions.
Maples and Calder has a leading commercial and litigation practice in the Cayman Islands, advising its clients on all aspects of public and private company mergers. In addition to acting for dissenters, Maples and Calder regularly represents respondent companies in connection with related appraisal and other proceedings.
The Grand Court's ruling in Re Shanda Games Limited ("Shanda") has been delivered.1 This is only the second merger fair value appraisal case that has gone to trial in the Cayman Islands,2 and the first that required the Court to determine the value of a company with assets and business operations in China. A copy of the full text of the case can be found here.
The Court concluded that the merger price (i.e. the amount offered to shareholders as consideration for the take-private) was too low. Following a forensic valuation process and a hearing that involved cross-examination of experts, the Court determined that the fair value of Shanda on the date that the merger was approved was more than double the amount offered to shareholders. This represents a significant victory for the dissenters. The 112 page decision of Segal J also advances the case law on the Cayman Islands merger regime following the ruling in Integra, and provides a further illustration of the Court's willingness to investigate the detailed and technical valuation issues that can arise in these cases. Maples and Calder acted for the successful dissenting shareholders in both cases, prevailing over Conyers Dill & Pearman in Shanda and Walkers in Integra.
Shanda, a Cayman Islands company listed on NASDAQ, was one of China's largest online games companies. It specialised in massively multiplayer online role-playing games and was developing its portfolio of PC and mobile internet games. Its American Depositary Shares ("ADSs") had listed in September 2009 at an offering price of US$12.50. Since then, the ADS price had declined and in late December 2013, it was trading at about US$5.50 per ADS. In January 2014, Shanda announced a take-private proposal with an indicated offer price of US$6.90 per share. Negotiations between the company, represented by a special committee of allegedly independent directors and their advisors, and the buyer group (which changed over time) continued for almost two years, finally leading to a concluded merger agreement and extraordinary general meeting ("EGM") in November 2015, at a price of US$7.10 per ADS. The buyer group controlled 90.7% of the total voting power at the EGM and the special resolution necessary to approve the merger was thus guaranteed.
Three ADS holders, which are all funds managed by Hong Kong-based fund manager Maso Capital, dissented from the merger and converted their ADSs to registered shares so as to be able to implement the appraisal process provided for under s.238 of the Cayman Islands Companies Law. Section 238 allows shareholders who wish to dissent from the merger to obtain a Court determination as to the company's fair value on the date of the EGM. Since the Integra ruling and the increase in the number of Chinese companies de-listing from US exchanges, the s.238 regime has become a feature of a number cases being dealt with in the Financial Services Division of the Grand Court: the majority of which involve companies with business operations and the bulk of their assets in China. Although a number of the other merger cases involving such companies have been the subject of interlocutory rulings or settlements, Shanda is the first such case to have gone to a full trial (Integra involved a Russian oilfield services company).
The s.238 dissent process requires the Court to determine the company's "fair value". That term is not defined in the Cayman Islands Companies Law. However, the merger regime follows similar statutory provisions in Canada and Delaware, where the courts have developed considerable expertise in dealing with the valuation issues raised by fair value appraisal cases.
Relying on numerous Canadian and Delaware cases, Jones J in Integra had concluded that the Court's role was to determine the fair value of the company's entire business as a going concern as at the valuation date, and that the fair value of a dissenter's shares would be their proportionate share of that amount, with no minority discount or premium for the forcible taking of their shares. There is no presumption that the merger price constituted a minimum price. This approach was adopted by Segal J in Shanda, who also cited leading Delaware authorities to confirm the proposition that the Court should use its judgement to evaluate the valuation opinions submitted by the experts, select the most representative analysis, and make appropriate adjustments to the resulting valuation or alternatively make its own independent valuation calculation by adapting or blending the factual assumptions of the parties' experts.
As is typical in the Grand Court, the Shanda case proceeded quickly. The case was commenced in February 2016, shortly after the expiry of the statutory time limits for the giving of notice of dissent and the mandatory negotiation period. Directions were given for disclosure and the exchange of expert reports. The dissenters made, and Shanda conceded to, an application for an interim payment of the merger consideration of US$7.10 per ADS.3 Following a number of interlocutory applications dealing with Shanda's obligations to produce documents and information, the trial took place within 12 months of the merger, in November 2016.
The format of the trial in Shanda was similar to that in Integra – namely with each side's expert producing a written report, contributing to a joint report, and then producing supplemental reports. Those reports were treated as the evidence of each valuer at the trial, and they were then cross-examined on their valuation opinions. Both experts had agreed that the most suitable valuation methodology was to solely use discounted cash flow ("DCF") models, and most of the cross-examination focused on the key components of the DCF models.
In Shanda the merger price, approved by the independent advisor at the time of the take-private as being "fair from a financial point of view", was US$7.10 per ADS. Following the disclosure and forensic assessment process, Shanda's own expert had concluded that this price was too low, and opined at the outset of the trial that fair value was in fact US$9.56 per ADS. However, following the cross-examination and legal submissions, Shanda conceded (before the trial finished) that the fair value was US$10.84 per ADS. The dissenters' expert concluded that fair value was far higher, at "a minimum of" US$27.16 per ADS. In its ruling, the Court made a number of rulings as to the disputed areas of valuation, and invited the parties to submit further calculations in order to determine the final amount of the ruling which will result in a final determination of fair value that is more than twice the merger consideration (we will provide a further update when the final number is determined). The Court's ruling deals with a number of the technical arguments typically associated with DCF valuation disputes. In resolving these questions and determining the fair value, the Court made a number of findings that will be of interest to participants in future merger cases.
In general, the way in which the fair value was determined by the Court is consistent with Integra and the approach taken in the Delaware and Canadian courts. Each company and each valuation is different. The Court will approach disputed issues on the basis that it is for both parties to establish, on the balance of probabilities, that the valuations their experts have presented are reasonable and reliable. If only one is reasonable and reliable the Court should generally follow that approach. If both appear to be reasonable and reliable then the Court must decide which is to be preferred. If neither is reasonable nor reliable, the Court must make its own determination. There were elements of all three of these options in the decision.
The Status of the Delaware (and Canadian) Authorities
In Integra Jones J had placed significant weight on the Delaware and Canadian case law on fair value appraisals.4 In Shanda, Segal J effectively endorsed Jones J's approach. While noting that the Delaware authorities are "obviously decisions of a different jurisdiction applied within a different procedural law and litigation culture", the Judge held that s.238 followed closely the model found in and "was drafted using the same core concepts and terms ... as appear in the law of Delaware (and Canada)". There is no similar merger legislation with analogous dissenter regimes in other jurisdictions to which Cayman law would typically look, including England. On that basis, and subject to the need to take care that the law and practice developed in those other courts remains consistent with other aspects of Cayman Islands law and practice (and also noting they are not binding), Segal J held that the Grand Court should have regard to and pay close attention to the Delaware and Canadian decisions.
No Minority Discount
In Integra, the parties had accepted, and the Court determined, that no minority discount should be applied to the dissenters' shares. However, Shanda argued that a minority discount should be applied on the basis that any market-based sale of the dissenters' ADSs would be subject to a minority discount. This argument was based on the fact that several English legal authorities dealing with unfair prejudice cases and buy-outs of a minority shareholder's interest were valued having regard to a minority discount. However, the Court concluded that the English authorities were distinguishable because the statutory language in England required a valuation based on a sale of shares, whereas the Cayman merger regime does not assume any such sale. Instead, Segal J preferred to adopt the position taken in Delaware where no minority discount is applied on the basis that what is being valued is the dissenters' interest in the corporation as a whole, rather than a particular block of shares. Fair value means the full interest of a dissenter's proportionate share of the capital and value of the company.
The DCF Analysis
Although the experts agreed that a DCF methodology was the only valid way to value Shanda, they differed on several key aspects of the DCF analysis which are likely to feature in similar future cases: (i) the discount rate and the inputs to its calculation; (ii) which of the three sets of projections produced by Shanda should be relied upon; (iii) whether a two-stage or three-stage growth model should be used; (iv) the treatment of expenditure on the take-private transaction; and (v) the treatment of restrictive stock units and share options triggered by the merger. There were also other disputes on case-specific features of the valuation (i.e. unique to Shanda's business and financial records and projections).
(i) Inputs to Discount Rate
The most significant component of the discount rate determined using the Capital Asset Pricing Model is often the beta calculation. Beta is an estimate of a company's risk as compared to the market as a whole and is determined by estimating the covariance of its returns against the market as a whole. Companies that are more volatile have higher betas. Because their cost of capital is higher, that adversely affects their value. A relatively small change in beta can substantially affect the discount rate.
Shanda's expert had decided to use Shanda's own beta (i.e. direct beta) and calculated it during the two year period prior to the merger date. Using monthly data, he concluded that Shanda's beta was 1.78. The dissenters' expert had concluded that Shanda's direct beta should not be used for a number of reasons, including that Shanda's expert's measurement period ended 22 months before the valuation date and used data within a period when the stocks of Chinese business companies listed on the US stock exchanges were being affected by a bias that affected risk assessment by investors. It was argued that the use of monthly betas was overly influenced by a single statistical measure, and inconsistent with the currency adjustment calculation made to the cash-flows. In addition, there was a dispute about the number of data points used and whether they were sufficient to form an accurate estimate of beta. The dissenters' expert argued instead that Shanda's beta should be calculated by reference to the betas of two comparable companies and using more data points, resulting in a beta of 1.001. The Court decided that there was merit in both approaches. The Court adopted the average of the experts' betas: 1.39, concluding that the data could and should be blended as between both the directly measured and indirectly observed betas. The Court noted that a similar approach had been taken in recent Delaware cases, notably In Re Appraisal of Dell Inc.5 and Re Appraisal of DFC Global Corp C.A.6
Equity Size Premium / Small Stock Risk Premium:
Small Stock Risk Premium ("SSRP") is often added to the company's cost of equity in the valuation of smaller companies to account for the higher rate of return demanded by investors to compensate for the greater risk associated with investing in small companies. Shanda's expert had proposed an SSP of 1.71%, by applying Shanda's market capitalisation to the commonly-used Ibbotson tables and in particular the decile categorisation. The dissenters' expert had instead calculated Shanda's value using a DCF approach and then applied that value to the Ibbotson small company SSRP of 1.07% - i.e. by using one of the three broader Ibbotson categories rather than the deciles. The Court preferred Shanda's approach.
Risk Free Rate, Equity Market Risk Premium and Country Risk Premium
Shanda had conceded during the trial to the dissenters' expert's view on Risk Free Rate, Equity Market Risk Premium and Country Risk Premium, so there were no findings by the Court on the basis for preferring a particular approach as to the selection of these inputs.
The combination of the Court's rulings on the disputed areas of beta and SSRP and the concessions made on the other factors enabled a calculation of the cost of equity (discount rate). This was done post-trial, with the parties agreeing the calculation and reverting to the Court for a final ruling. As the experts assumed no debt in Shanda's capital structure, the estimated cost of equity was the weighted average cost of capital.
(ii) The Projections
Shanda had prepared three different sets of projections and these were subject to investigation by the valuation teams. Shanda's expert conducted a high level review of the projections and was prepared to accept that Management's forecasting was reasonable, if somewhat "aggressive" based on the resemblance of the projection graphs to a hockey stick. The dissenters' expert scrutinised the projections and identified various apparent inconsistencies and errors, which the company could not fully explain in a rigorous pre-report question and answer process. Accordingly, the dissenters' expert made numerous corrections and adjustments to the projections. The suitability of the main corrections and adjustments was disputed.
The Court concluded that the dissenters' expert had conducted a very thorough review of the available evidence and had asked numerous questions about each of the main points, and had decided to make the corrections or adjustments based on many years of experience in the fields of accounting and valuation. In addition, in relation to one correction, the dissenters' expert had asked specific written questions of the company's management and had received cryptic and somewhat incoherent answers. On this issue, the Court noted that it would not generally be satisfactory to rely on such evidence. However, taking into account all of the evidence about the need for corrections and adjustments to the projections the Court was prepared to accept that corrections were indeed necessary. Furthermore, as the company had been given an opportunity to provide answers and explanations about the apparent inconsistencies and errors and had not done so, the Court could and should infer and conclude that the forecasts were unreliable. The Court concluded that the proposed amendments by the dissenters' expert were reasonable and realistic, and would be adopted.
This aspect of the judgment demonstrates one of the risks for companies that fail to engage properly in the Cayman Islands court process. Typically in fair value actions the company holds most or all of the relevant valuation information. A failure to provide proper disclosure and/or clear and direct answers to questions asked by experts in the lead-up to trial can cause difficulties for the company at trial. The Court may draw adverse inferences against the company when it comes to evidence or information it could have, but failed, to provide.
(iii) The Cashflow Model
Two or Three Stage Growth Model:
The experts differed on whether it was appropriate to use a two-stage or three-stage growth model. The dissenters' expert argued that a three-stage model (i.e. an explicit forecast period, a 10-year transitional period and then a terminal growth period) was appropriate in light of Shanda's high projected growth rates during the explicit forecast period. Shanda's expert argued that the company's projected growth rates were aggressive. The Court concluded that the company's forecast that growth would drop drastically from 18.5% in the last year of the explicitly forecast period to a terminal growth rate of 1 to 2% was not credible and that a three-stage model proposed by the dissenters' expert was appropriate. However Segal J also ordered that the transition period be reduced from 10 years to five years.
Terminal Growth Rate:
This Court's ruling reducing the length of the transition period led to the need to determine the growth rate in the terminal period starting after a five year rather than 10 year transition period, which had not been the subject of detailed evidence at the trial. The parties were required to make extra submissions as to the appropriate terminal growth rate which the Court will determine when granting its final order.
Foreign Exchange Rate:
The parties agreed to use a US$/RMB spot rate.
(iv) Treatment of Expenditure on Take-Private Transaction
This issue concerned the fees and expenses incurred by Shanda in obtaining advice concerning the merger and in implementing the merger itself. The dissenters' expert's view was that these should not be borne by the dissenters, whereas Shanda's expert thought that they should. The dissenters argued that the burdens of a transaction to which the dissenters did not agree – a forcible taking of their shares – should not be imposed on them. This argument was based in part on the influential Canadian case of Brant Investments v Keeprite7 in which the Court had stated "... the dissenting shareholders should not bear any diminution in the value of their investment directly attributable to the transaction from which they were entitled to dissent" and "... [the dissenting shareholders] should have no enhancement of the value in their investment attributable to the transaction which gave rise to their dissent." This had been referred to by Jones J in Integra, where he found that the percentage interest of the dissenting shareholders was to be fixed by the shares in issue at the valuation date, and should not be diluted thereafter. However, Segal J in Shanda was not prepared to extend this principle to take into account the costs and expenses connected with negotiating and implementing the merger agreement. As a consequence the dissenters were forced to pay their share of the costs of the merger.
(v) Treatment of Restrictive Stock Units
In Integra, Jones J determined that the dissenters' entitlement was to a fixed proportion of the issued shares on the valuation date, and was not to be diluted as a result of the issuance of further shares under share option agreements. Specifically, unvested stock units were to vest automatically upon the merger taking place. In Shanda, Segal J appears to have narrowed this principle: the Court ruled that if employees or others had rights to subscribe for and receive an issuance of shares that would be exercisable as at the valuation date without reference to or in reliance upon the merger, then a dissenter's proportionate shareholding would be subject to dilution by the issuance of such shares. However, if the right to an issuance of shares was conditional upon, and would only take effect upon, the closing of the merger, then the dissenting shareholder should not be affected by the issuance of such shares. In future cases, this will require a close analysis of the terms on which restricted stock units or employee stock options have been granted and are vested at the time of the merger.
The "Chinese re-listing" Point
One feature of the case was the observation by the dissenters' expert that empirical evidence strongly suggested that Chinese business companies listed on US stock exchanges traded at a discount due to wariness by US investors to the potential for corporate governance weaknesses. However, when re-listed on a Chinese stock exchange, many such companies traded at a significant multiple to the previous US listed price and the take-private price. Accordingly, the expert commented that the DCF-derived value was the "minimum" fair value because it might be reasonable to take into account the benefits that the buyer group might gain from the transaction in a Chinese re-listing.
As this observation did not form part of the expert's valuation conclusion, the Court was content to note it but concluded that the question of whether it would be permissible to rely for the purpose of a fair value assessment on the ability of a purchasing party to conduct a Chinese re-listing did not arise for determination in the case. The same was true of certain block transactions conducted between various factions of the buyer groups, which were noted as supportive of a higher value but not determinative of it, as both experts had reached their ultimate conclusions based solely on their DCF valuations.
The Shanda decision should give both prospective dissenters and companies a degree of comfort that, underpinning the Cayman Islands merger regime, is a reliable and (relatively) rapid process for the determination of fair value.
However, the decision should also serve as a warning that in all cases the Court will conduct a detailed review of each side's valuation analysis. In appropriate cases the outcome could be a significant uplift on the merger consideration (especially in the case of a take-private where the conflicts of interest for management participants will lead to scrutiny of all aspects of the valuation, including the budgeting and forecasting processes). As the detailed consideration given to these points in the Shanda decision makes clear, the result could be that corrections and adjustments to the inputs into the projections could have a significant effect on the DCF valuation output.
Unlike Integra, the production of the Court's written judgment might not immediately yield a number for the valuation, or a "price per ADS". Instead, the parties and their experts have been required to re-calculate after the trial, having regard to the Judge's decision on each of the DCF inputs. At the time of writing this analysis, the final price per ADS has not been publicly released, but will be more than double the merger consideration (to which interest will be added, and a further decision of the Court as to the appropriate interest rate is anticipated shortly). The dissenters will also seek an order for their costs of the proceeding. We will provide a further update when the final number is published.
1 In the Matter of Shanda Games Limited (FSD 14 of 2016, 25 April 2017, Segal J)
2 The first being the decision of Jones J in Re Integra Group  1 CILR 192 – see our updates First Ruling on "Fair Value" in the Context of Cayman Companies Mergers and Mergers in the Cayman Islands: Integra - Cost of "Fair Value" Assessment
3 This previously untested application by the dissenters to obtain an immediate payment of the merger consideration was subsequently confirmed in the recent ruling of Quin J in Re Qihoo 360 Technology Co Ltd (Unreported, Grand Court, FSD No. 129 of 2016, 26 January 2017, Quin J) – another Chinese-business merger case.
4 Generally speaking, and outside of this context, Delaware (and to a lesser extent depending on the circumstances, Canadian) authorities would not typically be regarded as persuasive in the Cayman Islands courts
5 C.A. No. 9322-VCL (Del. Ch. May 31, 2016)
6 C.A. No. 10107-CB (Del. Ch. July 8, 2016)
7 (1991), 3 OR (3d) 289 (CA)
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.