Deutsche Telekom c. Inde

This is an ad hoc arbitration brought under the Agreement between the Federal Republic of Germany and the Republic of India for the Promotion and Protection of Investments of 10 July 1995 (the "BIT" or "Treaty"),¹ pursuant to the 1976 United Nations Commission on International Trade Law ("UNCITRAL") Arbitration Rules (the "UNCITRAL Rules").

On 13 December 2017, the Tribunal issued an Interim Award (the "Interim Award") in which it decided that:

"a. The Tribunal has jurisdiction over this dispute involving the Claimant and the Respondent;

b. The Respondent has breached the fair and equitable treatment standard provided in Article 3(2) of the BIT;

c. The Tribunal will take the necessary steps for the continuation of the proceedings toward the quantum phase."⁷

"On the basis of the foregoing, without limitation and expressly reserving its right to supplement this request for relief, DT respectfully requests that the Tribunal:

(a) ORDER India to pay DT compensation in an amount of US$270 million for its breaches of the Treaty or such other sum as the Tribunal determines will ensure full reparation;

(b) ORDER India to pay pre-award interest on (a) above calculated at the rate of LIBOR plus 4% compounded annually from the Valuation Date until the date of the Tribunal’s award, or at such other rate and compounding period as the Tribunal determines will ensure full reparation;

(c) ORDER India to pay post-award interest on the same basis as pre-award interest accruing from the date of the award until payment is made in full;

(d) DECLARE that:

(i) the award of damages and interest in (a), (b) and (c) be made net of all Indian taxes; and

(ii) India may not deduct taxes in respect of the payment of the award of damages and interest in (a), (b) and (c);

(e) ORDER India to indemnify DT:

(i) for any taxes India assesses on the award of damages and interest in (a), (b) and (c); and

(ii) in respect of any double taxation liability that would arise in Germany or elsewhere that would not have arisen but for India’s adverse measures;

(f) AWARD such further or other relief as the Tribunal considers appropriate;

and

(g) ORDER India to pay all of the costs and expenses of this arbitration, including DT’s legal (external and internal) and expert costs, fees, and expenses, the fees and expenses of the Tribunal, the fees and expenses of any appointing or administering authority, the fees and expenses of any experts appointed by the Tribunal and the costs and expenses of any hearings (including the costs of DT’s witnesses in preparing for and/or attending such hearings), plus interest, pursuant to the discretion granted under Article 9(2)(b)(vii) of the Treaty and Article 40 of the UNCITRAL Rules."²⁰

"The Claimant relies on its Request for Relief as set out in its Reply on Quantum, subject to amending paragraph (b) such that interest is calculated at the rate of LIBOR (or any other comparable rate in case LIBOR should be discontinued in the future) plus 4% in line with the suggestion in paragraph 228 above."²¹

"For the reasons set forth above, the Tribunal should reject Claimant’s damage claim in its entirety and assess the costs of this case against Claimant."²²

i. the so-called "DT quantum papers" (i.e. DT’s Memorial on Quantum, supporting witness statements, expert report and documentary evidence) may be used in the Mauritius BIT Arbitration;

ii. the evidence presented in the Mauritius BIT Arbitration may be used in the present proceedings; and

iii. the Respondent may call as a witness in either case any witness who had submitted a witness statement on behalf of either the Mauritius claimants or DT in the quantum phase.

On 11 December 2018, the Tribunal was informed that the Swiss Federal Tribunal had denied the application to set aside the Interim Award.

a. For the Claimant:

• Sylvia Noury, William Thomas, Michael Kotrly, Annie Pan, Leonie Beyrle, Stephanie Mbonu and Madeleine Wall, of Freshfields Bruckhaus Deringer LLP, and Aman Ahluwalia;

• Claudia Bobermin and Ina Roth, Deutsche Telekom AG, Brian Thompson, Immersion Legal Graphics, Lau Nilausen and Mark Bosley, FTI Consulting ("FTI");

• Oliver Tim Axmann, Kim Kyllesbech Larsen, Gary Parsons and Axel Scheuermann, called as witnesses;

• Greg Harman, FTI Consulting, called as an expert.

b. For the Respondent:

• George Kahale III, Benard V. Preziosi Jr., Fernando Tupa, Simon Batifort, Fuad Zarbiyev, Gloria Diaz-Bujan and Abbey Li, Noémie Solle and Vincent Bouvard, of Curtis, Mallet-Prevost, Colt & Mosle LLP;

• Kalyani Sethurman, Misa, UR Rao Space Centre, Praveen Karanth and M S Krishnan, Department of Space, Prabeen Nair, Embassy of India, Paris, Krishna Mohan Arya, Ministry of Law and Justice, Raman Gupta, Enforcement Directorate, Ivan Vazquez, Quadrant Economics LLC;

• Bhagirath, Nitin Jain and Smt. M. Revathi, called as witnesses;

• Dr. Jacob Sharony, Mobius consulting, and Dr. Daniel Flores, Econ One Research Inc., called as experts;

• Pursuant to Article 3(c), Antrix was "responsible for obtaining all necessary Governmental and Regulatory Approvals relating to orbital slot and frequency clearances, and funding for the satellite to facilitate Devas services. Further, Antrix shall provide appropriate technical assistance to Devas on a best effort basis for obtaining required operating licenses and Regulatory Approvals from various ministries so as to deliver Devas services via satellite and terrestrial networks. However the cost of obtaining such approvals shall be borne by Devas";

• Further, under Article 12(a)(ii), Antrix, through ISRO/the Department of Space (the "DOS"), was "responsible for obtaining clearances from National and International agencies (WPC, International Telecommunications Union, etc.) for use of the orbital slot and frequency resources so as to ensure that the spacecraft is operated meeting its technical characteristics and provide the Leased Capacity as specified";

• Finally, according to Article 12(b)(vii), Devas was "solely responsible for securing and obtaining all licenses and approval[s] ([s]tatutory or otherwise) for the delivery of Devas Services via satellite and terrestrial network".

In the Terms of Appointment signed on 3 June 2014, the Parties agreed on the law governing the procedure of this arbitration as follows:

"40. In order of priority, the procedure in this arbitration shall be governed by the mandatory provisions of the law of the seat on international arbitration, these Terms of Appointment, the rules on procedure contained in Article 9 of the BIT and the 1976 UNCITRAL Arbitration Rules.

41. If the provisions therein do not address a specific procedural issue, the applicable procedural issue shall be determined by agreement between the Parties or, in the absence of such agreement, by the Arbitral Tribunal."⁵⁴

In respect of the law applicable to the merits, the BIT contains the following provision (Article 9(2)(b)(ii)):

"The arbitral award shall be made in accordance with the provisions of this Agreement, the relevant national laws including the rules on the conflict of laws of the Contracting Party where the investment dispute arises as well as the generally recognised principles of international law."⁵⁵

In this regard, it is the Claimant’s contention that the decision in Bilcon v. Canada⁶⁴ on which the Respondent relies does not support the latter’s argument that DT must prove "with virtual certainty" that it would have been granted the WPC License.⁶⁵ According to the Claimant, the finding in Bilcon hinged on "a pure lost profits valuation".⁶⁶ It is for that reason that the Bilcon tribunal held that the claimant was required to prove "with a certain level of probability" that it would have obtained the necessary regulatory approvals in order to establish that "future profits were indeed lost".⁶⁷ By contrast, in this case the loss sought is the FMV of the asset prior to the annulment of the Agreement. The licensing uncertainty in this situation goes to the amount of the loss, not to its existence. The Claimant finds support for this argument in the concurring opinion of Prof. Schwartz in Bilcon.⁶⁸

"Given the characteristics of the Ukrainian process for the awarding of licences, it is impossible to establish, with total certainty, how specific tenders would have been awarded if the National Council had not violated the FET standard. The best that the Tribunal can expect Claimant to prove is that through a line of natural sequences it is probable - and not simply possible -that Gala would have been awarded the frequencies under tender."⁷²

"the price, expressed in terms of cash equivalents, at which property would change hands between a hypothetical willing and able buyer and a hypothetical willing and able seller, acting at arm’s length in an open and unrestricted market, when neither is under compulsion to buy or sell and when both have reasonable knowledge of the relevant facts."⁷⁸

"In the Tribunal’s view, when quantifying the value of the expropriated assets, the Tribunal must proceed on the basis that Burlington is entitled to exercise all of the contractual rights it would have had but for the expropriation, and that Ecuador would have complied with its contractual obligations going forward. In other words, when building the counterfactual scenario in which the expropriation has not occurred, the Tribunal must assume that Burlington holds the rights that made up the expropriated assets and that those rights are respected."⁸⁰

a) the launch of the GSAT-6 satellite would have taken place in April 2011;

b) India would have acted "fairly and reasonably" in treating Devas’s application for the WPC License to permit terrestrial reuse of its spectrum; and

c) if needed, Devas would have acquired any service license in addition to the ISP and the IPTV licenses it already held.

Thus, for the Respondent, the core licensing issues in dispute are matters that need to be proven in order to establish a causal link between the breach of the Treaty and the loss.¹¹⁸ In assessing those core licensing issues, the applicable legal standard is one of "in all probability" or "with a sufficient degree of certainty". In support of this contention, the Respondent relies on Bilcon, where the tribunal held:¹¹⁹

"Authorities in public international law require a high standard of factual certainty to prove a causal link between breach and injury: the alleged injury must "in all probability" have been caused by the breach [...] or a conclusion with a "sufficient degree of certainty" is required that, absent a breach, the injury would have been avoided.

[...]

In this regard, the test is whether the Tribunal is "able to conclude from the case as a whole and with a sufficient degree of certainty" that the damage or losses of the Investors "would in fact have been averted if the Respondent had acted in compliance with its legal obligations" under NAFTA."

The Respondent stresses that Bilcon applies to "all aspects of the Tribunal’s finding of liability, including non-transparency in the decision-making process and the annulment itself".¹²⁰ In this regard, India draws the Tribunal’s attention to the fact that in Bilcon the tribunal "went on to note that various outcomes of a ‘NAFTA-compliant process’ were ‘reasonably conceivable’, including a refusal to approve the project based on socio-economic considerations and an approval ‘with conditions that would render [the project] economically unviable’".¹²¹ The Bilcon tribunal thus concluded that "[w]ithout a high degree of certainty as to regulatory approval, it goes without saying that no damages based on the profitable operation of the quarry can be awarded".¹²² Contrary to the Claimant’s submission, according to the Respondent, there is "no meaningful distinction between Bilcon and this case".¹²³

a) As DT was aware, India had no obligation to grant the WPC License and DT should not be able to claim for a right it never had (infra at V.A.2.b(i));

b) DT fully understood that there was significant risk that the licenses would not be granted (infra at V.A.2. b(ii)-(iii)); and

c) DT has failed to prove that the spectrum fees would have been set at a level that would have enabled Devas to be profitable (infra at V.A.2.b(iv)).

• DT’s internal documents record DT’s reservations about the grant of the license, including after DT’s officials met with India’s Wireless Advisor to discuss this matter;¹⁴²

• The uncertainty surrounding the grant of the WPC License was factored in by DT itself when it invested in Devas, as is shown by (i) DT’s decrease of the valuation of Devas and its proposal for a staged investment, and (ii) the fact that it obtained priority shares ensuring that in the event of a liquidation or sale of the company, DT Asia would receive preferential treatment vis-à-vis holders of ordinary shares;¹⁴³

• In the lead up to its second investment in Devas, DT continued to take note of the licensing risks.¹⁴⁴

• As confirmed by Ms. Revathi, the level playing field policy "means exactly what it says - that companies using the same (or similar) spectrum or desiring to provide the same types of services are required to endure the same fees and charges, and the Government is not permitted to regulate in a manner that would give one an advantage over the others [...]";¹⁴⁹

• Second, "no one in India provides BWA services, and no one has ever been authorised to provide BWA services, without paying auction prices", even existing service providers would have been required to pay a fee to upgrade their services "commensurate with auction prices".¹⁵⁰ The granting of such a license would have required a fundamental change in Indian regulatory policy and the evidence before the Tribunal is that such a change would not have taken place;¹⁵¹

• Third, the July 2010 DOT Letters made it clear that spectrum charges in line with auction values would be assessed for the proposed satellite-terrestrial business.¹⁵² There is no reason to disregard this letter, as there is nothing unlawful about the Secretary of the DOT reminding the DOS that, in accordance with the level playing field policy, the spectrum charges would have to equate auction fees.¹⁵³

There is no serious disagreement between the Parties as to the applicability in the present case of the standard of full reparation set out by the PCIJ in its 1928 judgment in Chorzów.¹⁶² The same standard is found in Article 31 of the International Law Commission Articles on State Responsibility (the "ILC Articles") and adopted by many investment treaty tribunals.¹⁶³

The Tribunal begins its analysis by recalling that, in accordance with Article 31 of the ILC Articles, the determination of damages under international law implies a three-step process:

i. establishing a breach;

ii. ascertaining that the injury was caused by that breach (causation); and

iii. determining the amount of compensation due for the injury caused (valuation or quantification of damages).¹⁶⁴

The first step was taken in the Interim Award, in which the Tribunal held that India’s conduct, ultimately culminating in the Devas Agreement being annulled, constituted a breach of the fair and equitable treatment ("FET") standard.¹⁶⁵ In particular, the Tribunal found that "the decision to annul the Agreement was arbitrary and unjustified inasmuch as it was manifestly not based on facts, but on conclusory allegations, and was the product of a flawed process".¹⁶⁶ The Tribunal concluded that "[a]s a consequence of the acts which the Tribunal deemed contrary to FET, the Devas Agreement was annulled".¹⁶⁷

The second step requires showing a causal link between the breach and the alleged injury. In this respect, the Tribunal agrees with the tribunal in Bilcon that "[a]uthorities in public international law require a high standard of factual certainty to prove a causal link between breach and injury: the alleged injury must ‘in all probability’ have been caused by the breach (as in Chorzów), or a conclusion with a ‘sufficient degree of certainty’ is required that, absent a breach, the injury would have been avoided (as in [the] Genocide [case decided by the ICJ])".¹⁶⁸

Bearing in mind the Chorzów and Genocide standards of proof mentioned above,¹⁷¹ in the Tribunal’s view, the Claimant has thus established that the loss (i.e. the diminution of value of its investment in Devas) has "in all probability" (pursuant to the Chorzów standard) or to "a sufficient degree of certainty" (pursuant to the Genocide standard) been caused by India’s conduct.¹⁷² Hence, the Tribunal considers that causation between the breach and the (fact or existence of the) loss has been established.

The Tribunal turns now to the third step, i.e. the valuation or quantification of damages. As also recalled in Bilcon,¹⁷³ investment treaty case law draws a distinction between the causation aspect just examined and quantification of the amount of the loss, for which a lower standard of proof is required.

The tribunal in Lemire, for instance, noted that:

"[o]nce causation has been established, and it has been proven that the in bonis party has indeed suffered a loss, less certainty is required in proof of the actual amount of damages : for this latter determination Claimant only needs to provide a basis upon which the Tribunal can, with reasonable confidence, estimate the extent of the loss."¹⁷⁴

Other tribunals have come to the same conclusion. For instance, the tribunal in Crystallex v. Venezuela, referring to Lemire, held:

"the fact (i.e., the existence) of the damage needs to be proven with certainty. In that sense, there is no reason to apply any different standard of proof than that which is applied to any other issue of merits (e.g., liability).

Second, once the fact of damage has been established, a claimant should not be required to prove its exact quantification with the same degree of certainty. This is because any future damage is inherently difficult to prove."¹⁷⁵

As the tribunal in Vivendi II further observed in respect of quantification, "approximations are inevitable; the settling of damages is not an exact science".¹⁷⁶ Tribunals thus "retain a certain margin of appreciation" in the quantification of the damage, as noted by the tribunal in Rusoro v. Venezuela.¹⁷⁷

This point was clearly articulated in Gold Reserve v. Venezuela, in which the tribunal held:¹⁷⁸

"while a claimant must prove its damages to the required standard, the assessment of damages is often a difficult exercise and it is seldom that damages in an investment situation will be able to be established with scientific certainty. This is because such assessments will usually involve some degree of estimation and the weighing of competing (but equally legitimate) facts, valuation methods and opinions, which does not of itself mean that the burden of proof has not been satisfied. Because of this element of imprecision, it is accepted that tribunals retain a certain amount of discretion or a "margin of appreciation" when assessing damages, which will necessarily involve some approximation. The use of this discretion should not be confused with acting on an ex aequo et bono basis, even if equitable considerations are taken into account in the exercise of such discretion. Rather, in such circumstances, the tribunal exercises its judgment in a reasoned manner so as to discern an appropriate damages sum which results in compensation to Claimant in accordance with the principles of international law that have been discussed earlier."

Hence, once it is proven that the Claimant did suffer a loss, and that loss was caused by the breach, the question is whether the Claimant has "provided the Tribunal with a reasonable basis to assess" that loss¹⁷⁹ or "a basis upon which the Tribunal can, with reasonable confidence, estimate the extent of the loss".¹⁸⁰

Still in respect of risk, the Tribunal observes that it is correct that, as the Claimant submits, the applicable counterfactual must assume that India would have acted in accordance with its Treaty obligations.¹⁸² It also does not escape the Tribunal that India’s breach deprived Devas of the opportunity to apply for the WPC License. However, this does not mean that the Tribunal should disregard the risks or uncertainties facing Devas irrespective of the Respondent’s breach. Hence, for purposes of valuing the Claimant’s loss, it is important to assess the risks, especially the regulatory risks on which the Parties have mainly focused, that Devas and its business would have faced even if the Agreement had not been annulled.

"In addition to various commercial risks, such as the greenfield nature of the project and the satellite aspect of the platform, the board [of DT] was fully advised of key legal and regulatory risks, including the need for a licence from the WPC authorising terrestrial reuse of the satellite spectrum, which had never before been issued or considered in India."¹⁸⁴

"MR PREZIOSI: Now, you did understand [...] this was a novel set of services in India, right?

A. In India, yes. That was our understanding.

Q. It’s correct, isn’t it, that at that time nobody had been providing AV broadcasting services to mobile devices in India?

A. In India, I don't believe so.

Q. And nobody, in fact, had been providing AV broadcasting to mobile devices terrestrially in India?

A. I would be able to confirm that. But it was a very novel concept, that was our understanding, yes."¹⁸⁵

"[l]n a meeting with the head of WPC in Delhi, it became apparent that Devas’ assumption that is has secured a substantial spectrum via its contract with ISRO may not go unchallenged by authorities other than ISRO, with the WPC-Chairman indicating a need for further review."¹⁸⁶

"License Requirements (cont’d):

Possible scenarios in the given context are:

A. WPC adopts the Devas position and takes the view that only procedural/technical clearances are required by Devas, possibly along with the payment of a fee for use of spectrum.

B. WPC concludes that the regulatory framework is unclear and that before it can come to a decision, relevant rules need to be framed in connection with the use by private operators of satellite spectrum in conjunction with a hybrid network.

C. WPC takes the view that Devas is required to obtain a substantive authorization from WPC in connection with the use of spectrum.

Scenario B. and C. would result in a substantial risk that the deployment of the Devas network would get delayed, and perhaps even refused in certain circumstances.

Other points of concern are that more generally, India has to be considered as a highly regulated market, and it is difficult to foresee how it will evolve over the near-to long-term."¹⁸⁷

"[MR AXMANN] Well, now that you mention it, scenario 1 and scenario 3 especially were given small probability. Indeed, our highest assumed probability was that it would have to run through a certain process, but it would be agreed then in the end.

Q. Okay. So just to be clear, A and C were given very small probability by DT, and scenario B was the one that you were giving higher probability to?

A. From my perspective and from the team's perspective, yes."¹⁸⁸

"DT Position: The fundamental concern to DT in the given context is that Indian regulations explicitly stipulate that any ISP-licensee is required to obtain ‘a separate specific authorization and license [...] from the WPC’, as the responsible agency for spectrum licensing and management. In addition, guidelines for Telecom Service Providers for Satellite Communications require the licensee to approach the WPC to obtain frequency authorization."¹⁹⁰

"To get clarity on the matter, DT (Kevin Copp in person, Hamid Akhavan via telephone) did meet/talk with the Chairman of the WPC. However, the feedback was non-committal. Accordingly, DT requested to eliminate any uncertainties by way of confirmatory letter either from WPC directly or from ISRO/DoS, explicitly confirming either the approval from, or the non-responsibility of WPC. This has not been obtained so far and Devas has indicated that, at least at this stage, it is reluctant to approach the authorities with the request for a formal clarification."¹⁹¹

"Spectrum: A risk arises out of the questions surrounding the competencies of various Indian authorities in connection with the exceptionally valuable spectrum assigned to Devas. An exclusively satellite-based use of the spectrum is guaranteed by the competence of ISRO; at least in an exclusively terrestrial use of spectrum, on the other hand, a regulatory approval by the Department of Telecommunications (DoT) would be necessary. It cannot therefore be excluded that the DoT could infer a competence in case of a subsequent terrestrial expansion.

[...]

Explicit authorisations for the terrestrial use can however only be obtained once the expansion of the network renders this necessary. Up to this point it cannot be completely excluded that there will be complications. On the other hand this risk element causes a lower valuation (see also the section ‘DT investment activity’)."¹⁹⁵

In sum, on the Valuation Date there was an element of uncertainty as to (i) the principle of the WPC License being granted to Devas and (ii) the amount of the fee payable for such grant. This conclusion does not mean that Devas was worth zero on the Valuation Date, as the Respondent contends. Rather, the Tribunal must adopt a valuation methodology which takes account of the specific circumstances of this case and the evidence on valuation in the record, without disregarding the level of uncertainty on the licencing issues. In other words, such uncertainty is but one of the elements which the Tribunal may consider in selecting the valuation methodology that, based on the evidence in the record, provides the most appropriate "basis upon which the Tribunal can, with reasonable confidence, estimate the extent of the loss".²¹⁰

The Claimant explains that the DCF method is the most appropriate in the present case for a number reasons:

• It is a valid method for evaluating start-up businesses, particularly where it is necessary to adjust for certain risks such as the licensing risks present in this case;²¹⁵

• It is well-established for the valuation of businesses in the telecommunications sector generally and within DT more particularly;²¹⁶

• DT has applied the DCF method in 2008 and 2009 in the ordinary course of business, when it made its investment in Devas;²¹⁷

• In February 2011, it was a requirement of India’s currency laws that a foreign buyer purchasing shares in an unlisted company, such as Devas, value the company using the DCF method;²¹⁸

• Devas’s primary asset, i.e. its right to lease the spectrum under the Agreement, was an income-producing asset and DCF is the only reliable method of assessing such an asset, as any other method would dramatically undervalue the asset;

• Arbitral tribunals have endorsed the use of the DCF method for the purposes of valuing a telecommunications company, even when the company was not a going concern. DT points in particular to the decision in Rumeli v Kazakhstan,²¹⁹ and

• It is a well-established principle "that the FMV of an asset must be assessed by reference to its ‘highest and best use’".²²⁰

In this regard, the Claimant contends that the absence of a track record of profitability does not automatically render the use of the DCF method overly speculative, if the investor can prove that the investment would have generated the profits claimed.²²² In this regard, the Claimant cites to Vivendi II, which held that:

"a claimant might be able to establish the likelihood of lost profits with sufficient certainty even in the absence of a genuine going concern" by "presenting sufficient evidence of its expertise and proven record of profitability of [businesses] it (or indeed others) had operated in similar circumstances."²²³

• Devas was well-advanced in the preparations to launch its operations. It had prepared a network roll-out plan, identified and priced network components, coordinated discussions with third party buyers, developed hand held terminals and devices, performed experimental trials, developed societal applications, and prepared the WPC License application;²²⁸

• Even though no other company had a track record of operating similar services in India, the BWA services that Devas was to offer were the "bread and butter" of DT, and for the AV services, Devas benefitted from some of the world’s foremost experts in hybrid satellite-terrestrial systems. Moreover, such services and businesses were "well known to Devas and DT" outside of India;²²⁹

• The "Darwin Model", i.e. the most up to date version of DT’s business plan developed with Devas in September 2009, "which FTI has taken as the starting point for its valuation", was developed in the ordinary course of business, not in the context of a dispute, and contained robust assumptions in relation to the Devas business;²³⁰

• DT rebuts India’s argument that its valuations played no role in the price it paid for its shares in Devas by stating that its contemporaneous valuations set a theoretical "ceiling" on the price DT would have paid for a business like Devas. The fact that DT was ultimately able to negotiate the price down does not affect its "real-time" view of the value of Devas;²³¹

• Contrary to India’s allegations, the fact that DT’s 2008 financial statements did not include the cash flow forecasts set out in the 2008 business plan, does not mean that DT considered those cash flows unreliable. Valuations performed for accounting purposes are not the same as valuations prepared for the purposes of corporate transactions,²³² and "the accounting team would not have reviewed Devas’s cash flows".²³³

For India, the authorities are "virtually unanimous" in their conclusion that DCF is inappropriate in circumstances such as the present ones.²⁵⁸ Notably, the World Bank Guidelines on the Treatment of Foreign Direct Investment, the ILC Articles Commentary, as well as a number of authorities and investment treaty awards, have rejected the use of the DCF method to value assets that lacked an established record of profits, on the ground that the contrary approach would result in speculative valuations.²⁵⁹ India further underscores that the ICC tribunal also discarded the DCF method.²⁶⁰

It is the Respondent’s submission that the decisions relied upon by DT in support of its argument that the DCF method was applied in other disputes involving the telecommunications industry, even when the company was not a going concern, are all inapposite based on the particular facts of those cases.²⁶¹ Those decisions reinforce the point that there is "no precedent of a tribunal using DCF under these circumstances, or anything remotely resembling them, absent agreement of the parties".²⁶²

• In the absence of the critical WPC License, the Claimant is wrong to suggest that "Devas’s primary asset, being its right to ‘the lease of valuable satellite spectrum’ under the Agreement, was an income-producing asset";²⁶⁴

• DT’s reliance on its DCF computations in 2008 and 2009 is misconceived, as those valuations played no apparent role in setting the price for the acquisition of the Devas shares,²⁶⁵ nor were they updated to reflect fundamental changes in the market by the time of the Valuation Date;²⁶⁶

• DT admitted in its own 2008 Annual Report that it "did not measure the investments by discounting the expected cash flows because the cash flows could not be reliably determined".²⁶⁷ Even if, as the Claimant asserts, there is a difference between the valuation method used for corporate as opposed to accounting purposes, that does not explain why the cash flows could not be reliably determined, or how the language used "does not in fact reflect an opinion on the Devas cash flows".²⁶⁸

Fourth, the Respondent asserts that the 19% discount rate used by the Claimant is speculative and does not accord with the discount rate applied to a start-up. According to the literature, the average discount rate for a start-up such as Devas is in the order of 30-70% and Dr. Flores has set the correct rate here at 31%.²⁷⁵ India further contends that the discount rate of 20% which DT used in its internal DCF computations did not consider licensing risk, as was confirmed at the Hearing by Mr. Axmann, the Vice President of DT’s Mergers & Acquisition team and then project manager in charge of evaluating the Devas investment opportunity.²⁷⁶

"The first step in valuing an asset pursuant to the DCF method must be to project from the valuation date onward the most likely revenues and expenses of the ongoing concern, year by year. The revenues less the expenses will give the future cash flow. The second step will be to discount the projected net cash flow to its ‘present value’ as of the valuation date."²⁷⁷

"Normally, the fair market value of a going concern which has a history of profitable operation may be based on an estimate of future profits subject to a discounted cash flow analysis.

However, where the enterprise has not operated for sufficiently long time to establish a performance record or where it has failed to make a profit, future profits cannot be used to determine going concern or fair market value. [...]

The Tribunal agrees with Mexico that a discounted cash flow analysis is inappropriate in the present case because the landfill was never operative and any award based on future profits would be wholly speculative."²⁸⁰

Numerous arbitral tribunals, including those in Phelps Dodge v. Iran,²⁸¹Southern Pacific Properties v. Egypt,²⁸²Wena v. Egypt,²⁸³ and Tecmed v. Mexico,²⁸⁴ among others,²⁸⁵ have adopted a similar reasoning. The arbitral tribunal in Siag v. Egypt, for example, referred to "the wisdom in the established reluctance of tribunals [...] to utilise DCF analyses for ‘young’ businesses lacking a long track record of established trading", which reluctance, it said, "ought to be even more pronounced in cases [...] where the business is still in its relatively early development phase and has no trading history at all".²⁸⁶ In some of these cases, even where the production or business activity had already started, tribunals nonetheless declined to award damages based on forecasts of future cash flows on the ground that the track record was deemed insufficiently reliable.²⁸⁷ The Tribunal agrees with this well-established line of cases and considers that this jurisprudential trend is not, contrary to what the Claimant appears to suggest, outdated, but includes several recent examples, such as Caratube v. Kazakhstan²⁸⁸ or South American Silver v. Bolivia.²⁸⁹

With those principles in mind, the Tribunal considers whether DCF would be appropriate in light of the reality of the Devas business. It is common ground that Devas was not a going concern. Its proposed business had not started, it lacked any customers, its cost levels were untested, and it had not yet generated any revenues. It thus had no track record of profitability whatsoever. In the Tribunal’s view, these facts would suffice in and of themselves to discard DCF as an appropriate valuation methodology. This conclusion is reinforced by the fact that Devas lacked the WPC License,²⁹² the issuance of which was uncertain on the Valuation Date (see supra section V.A.3.b), as was the level of the related fee.

"At the balance sheet date, T-Mobile Venture Fund GmbH & Co. KG and Deutsche Telekom Asia Pte Ltd were recognized at cost. No market prices were available for the investments. Neither was it possible to derive the respective fair value in the period in question using comparable transactions.

The Company did not measure the investments by discounting the expected cash flows because the cash flows could not be reliably determined."²⁹³

Finally, the Tribunal finds confirmation for its conclusion that a DCF valuation is inapposite in the decision of the ICC Tribunal in Devas v. Antrix, which viewed the DCF methodology presented by Devas as "an unrealistic and unreliable vehicle for determining its damages":

"[T]he demand for Devas’ services is unclear; the prices that it would be able to profitably charge is unclear; market(s) for multimedia broadcasting services can be highly innovative and cause (even very profitable) products and services to quickly become obsolete; and there is persuasive evidence [...] that Devas faced significant competition for the services that it proposed to provide. In other words, there is nothing that can give the tribunal sufficient confidence about the cash flows that Devas would have [...]. Further, an aspect of the DCF methodology that the tribunal finds particularly troubling in this case is that small variations in the assumptions used in the DCF methodology can dramatically and unrealistically change Devas’ value [...] The tribunal understands that the reason for the extreme sensitivity of the DCF methodology in this case is the length of the period that it would take for Devas to become cash flow positive (nine years). In this case, in the tribunal’s view, it makes Devas’ DCF methodology an unrealistic and unreliable vehicle for determining its damages."²⁹⁴

In conclusion, the lack of operating history, customers and profitability and the relatively early stage of the project lead the Tribunal to the conclusion that the DCF method cannot form the basis for the quantification of the Claimant’s damages. This conclusion takes into account but does not view as determinative the uncertainty regarding the WPC license which would have been necessary to roll out the proposed services and, therefore, generate profits. Accordingly, the Tribunal discards the DCF valuation put forward by FTI which, in light of the objective factors just mentioned, would be subject to excessive uncertainties, contingencies and hypotheses, and would not provide "a basis upon which the Tribunal [could], with reasonable confidence, estimate the extent of the loss".²⁹⁷

a. It takes as a starting point DT’s March 2008 cash payment of USD 75 million for its investment in Devas, which is said to imply a valuation of USD 375 million for Devas at that time;

b. It adjusts that amount to reflect Devas’s FMV by factoring in DT’s in-kind contributions and bargaining power; and

c. It adds an uplift fee to reflect the progress made in developing the business between March 2008 and the Valuation Date.³⁰⁰

a. The risks associated with investing in Devas, including the WPC Licensing uncertainty;³⁰²

b. DT’s negotiating power and leverage due to (i) the significantly lower price paid by the two venture capital funds, which acquired shares in Devas in March 2006 and 2007; (ii) the fact that Devas had relatively limited funding needs prior to the satellite launch; (iii) DT’s assumption that it was the only interested investor offering Devas strategic benefits; (iv) DT’s knowledge that, should Devas be required to search for an alternative investor, it would have incurred substantial delays;³⁰³ and (v) the fact that Devas was under a compulsion to sell due to its need to secure a strategic investor by early 2008 in order to ensure the roll out of the business;³⁰⁴

c. DT’s in-kind contribution, including the fact that Devas was a strategic investor bringing "technical expertise and considerable procurement leverage".³⁰⁵ The Claimant contends that Devas would not have accepted an investment on the same monetary terms from another investor.³⁰⁶ This argument is supported by the fact that from March 2008 Devas did indeed benefit from DT’s operational and technical expertise as well as "sourcing and procurement support [...] [of] network equipment at a considerably lower cost";³⁰⁷ and

d. DT’s concerns about its rights as a minority shareholder, particularly considering the potential for diverging interests between itself as a strategic investor and Columbia Capital and Telecom Ventures as financial investors. For that reason, DT’s usual practice was to take a controlling stake in its subsidiaries.³⁰⁸

Table 6-2: Components of the 2008 Adjustment

a. The fact that DT held Class C priority shares does not mean that its investment in Devas in 2008 cannot be used as a starting point for Devas’s value. While Devas had five classes of shares, DT assumed that there was no economic value in Class E shares and that Class D shares had the same economic value as Class A, B and C shares. On that basis, it considered its 17% of issued share capital was equivalent to a 20% economic stake in Devas;³¹¹

b. DT did not seek to protect itself from licensing risks by way of representations and warranties.³¹² India’s arguments in this regard misunderstand and magnify certain comments found in documents. If DT sought such representation and warranties, they would appear in the 2008 and 2009 Share Subscription Agreements or in the papers presented to the DT’s Management or Supervisory Boards in 2008 and 2009;³¹³

c. India’s reliance on the fact that one of the claimants in the Mauritius Shareholders Arbitration paid only USD 146 per share to acquire its ordinary shares in September 2009 is similarly unfounded. That purchase was made through Devas Employees Mauritius Private Limited ("DEMPL"), a "vehicle by which Devas management and other employees could acquire shares in Devas as part of their compensation and rewards package pursuant to an Equity Incentive Plan".³¹⁴ The share price for that transaction is not relevant to DT’s investment in Devas.³¹⁵

FTI’s assessment on reasonable weighting.³¹⁶ The values adopted by FTI for the bargaining power and in-kind contribution are conservative and supported by the following facts:

a. Devas’s initial proposal for the sale of the shares implied a pro rata value of USD 800 million, whereas the price paid by DT implies a pro rata value of USD 375 million. This demonstrates that, using its "take-it-or-leave-it" approach (among other things), DT was able to achieve a reduction of USD 425 million, which is in fact more than the USD 370 million FTI has attributed to bargaining power;³¹⁷ and

b. Based on the DCF model used by DT for the initial investment (i.e. the Series C 2008 Model) "DT’s support and procurement power would only have needed to reduce Devas’s expected operating and capital costs by circa 2.9% to increase Devas’s value by USD 221 million",³¹⁸ a fact confirmed by Dr. Larsen.³¹⁹ In this regard, India’s reliance on the evidence of Mr. Scheuermann and the allegation that the "in-kind" contribution was not recorded in any of the contemporaneous Board papers or documents is both incorrect and misplaced.³²⁰ This is because:

■ Mr. Scheuermann’s role was to conduct DCF valuations, whereas Dr. Larsen was "responsible for the in-kind element of the investment" and Mr. Axmann "carried out that price negotiation".³²¹ Moreover, Mr. Scheuermann’s evidence concerning DT’s practice of accounting for in-kind contributions by a "synergy calculation spreadsheet", is confined to calculations "between two established telecom companies".³²²

■ The Board was informed that DT was to provide "dedicated resources in the area of Procurement, Terrestrial Network Design & Planning Deployment",³²³ both in the form of "manpower" (which Dr. Larsen estimated at USD 6-7 million) and in terms of procurement power and CapEx cost savings.³²⁴

■ Finally, "it would not have made sense to update the Devas business plan to provide for the full extent of the procurement benefits that had not yet been negotiated, much less obtained".³²⁵

a. Increasing the value of Devas by 25%, rather than 50%, to account for the fact that the ICC Tribunal may have factored the "in-kind" contribution into its calculation when using 50%;³²⁹ or

b. Increasing the value of Devas by adding the uplift applied by the ICC Tribunal in absolute terms only (i.e. 50% of USD 375 million = USD 187.5million) to account for the fact that "it is impossible to determine whether the [ICC] Tribunal had a relative or absolute increase in mind when it effected a 50% uplift".³³⁰

a. The fact that DT paid the same price for its shares in 2008 and 2009 is not indicative of a decrease in Devas’s value. The reason why the price was identical is "primarily as a result of the global financial crisis".³³² In fact, contrary to India’s view, DT’s willingness to pay the same price despite the financial crisis demonstrates that Devas had increased in value;³³³

b. There is no evidence suggesting an increase in risk concerning the WPC License.³³⁴ This is buttressed by the fact that the ICC Tribunal concluded that the lack of WPC License had no adverse impact on the value of Devas;³³⁵

c. The reliance which India places on the 2010 BWA Spectrum Auction as evidence of increased WPC License risk is misplaced. The evidence demonstrates that "the high prices likely to be paid in the BWA auction were value-enhancing, since they were viewed as a proxy for the value of the spectrum rights owned by Devas, not the cost of the WPC License";³³⁶

d. Dr. Flores is wrong to suggest that satellite risks increased prior to the Valuation Date. Progress was made on constructing the satellites and Devas considered alternate launch vehicles. The slow roll-out of the first launch should be disregarded on the basis that by July 2010 India had decided to annul the Devas Agreement;³³⁷ and

e. Prior to the Valuation Date, there were favorable technological and economic developments that reduced risks inherent in the project, such as development in TD-LTE for the roll out of Devas’s BWA business, and improvements in macroeconomic factors.³³⁸

a. DT invested USD 75 million in preferential shares and on the basis of representations and warranties of Devas. It is therefore wrong to conclude that Devas’s value was at least USD 375 million in 2008;

b. The USD 600 million uplift for in-kind contributions and bargaining power is unjustified;

c. The further upward adjustment of USD 483 million on the ground that the value of the Devas business increased from 2008 to 2011 is also untenable.³⁴⁰

a. Following the BWA Auctions, as part of its due diligence, a hypothetical buyer would have sought clarification from the regulator and would have been informed that there was no existing licensing regime permitting Devas to provide the services envisaged. Similarly, the hypothetical buyer would have been advised that "even if the required licenses had been issued, Devas would have had to pay fees commensurate with the 2010 auction values, as reflected in the July 2010 letters written by the highest officers of the DOT and WPC";³⁵⁴

b. A hypothetical willing buyer would have been alerted by delays in launching the GSAT-6 satellite, which is supported by Devas’s concern about increasing delays.³⁵⁵ Contrary to the Claimant’s arguments there is simply no evidence that the launch was deliberately "slow-rolled".³⁵⁶