The conversation about third-party funding in international arbitration always seems to focus on claimant-side products. In this paper, I will instead introduce two insurance products available to respondents: (1) "after-the-event" (ATE) insurance for legal costs; and (2) "outcome hedging", "caps" or "litigation buyout" insurance to create a stop-loss on liability (outcomes policies). Both of these products have significant virtues but also significant vices, including high premiums and limited availability (unlike ATE insurance for litigation in English courts). But the international arbitration community may benefit from a broader discussion about employing these risk management tools. Nevertheless, respondent-side risk protection products are uncharted waters for international arbitration. Navigators should beware of the unexpected.

1 "AFTER-THE-EVENT" (ATE) INSURANCE

ATE insurance covers a respondent policyholder against the potential costs of legal action brought against that respondent by an opponent. If the insured loses the dispute and the opponent's legal costs and expenses are apportioned to the insured respondent, then the insurance company will pay those legal costs and expenses. In addition, ATE insurance will cover the policyholder's own out-of-pocket disbursements. A disputing party ordered by the tribunal to deposit security for costs in a proceeding may also seek to deposit an ATE insurance policy in satisfaction of the order, rather than cash.

ATE policies are well-known to English litigation practitioners. Undoubtedly due to the organized application of cost-shifting for legal fees and expenses in the English judicial system, a vigorous market for ATE insurance products exists in Britain to provide economic support for unsuccessful litigants who thereby become responsible for the fees and costs of the victorious party.

One prominent British law firm has estimated the level of premiums for claimants in English litigation at 30-50% of the sum insured.1

level of premium payable depends on the type or level of cover sought and assessment of the risk, can typically be 30% to 50% of the sum insured, or may be calculated as a percentage of the costs incurred at the date a claim is successfully concluded by negotiation and/or in court proceedings."

Insurers have offered the product based on upfront premiums, staged premiums and even premiums payable solely out of a successful outcome. Like any insurance product, ATE insurance is customizable, but certain features discussed below regarding inter alia limits and exclusions are easily predicted.

Until recently, the successful litigant in English court proceedings could recover ATE insurance premiums from the losing litigant, in addition to recovering legal fees and expenses. However, an influential report by Lord Justice Sir Rupert Jackson released in early January 20102 is leading to regulatory changes in English courts with respect to that practice in 2013.

ATE insurance in England is "usually for claimants"3 but is by no means unavailable to respondents. To a considerable extent, the English insurance market had heretofore configured its ATE offerings to charge a premium payable only if a successful outcome for the insured litigant had occurred, and the premium was then recoverable under English cost-shifting principles from the losing party. That "recoverability" framework has been abolished by the Legal Aid, Sentencing and Punishment of Offenders Bill,4 and related regulatory enactments, coming into effect in April 2013. As a consequence, English insurers are restructuring their ATE products. Higher premiums are likely in the English market.

The controversy in Britain over the recoverability of ATE premiums from the losing party by a successful party that purchased ATE insurance has focused on the use of the product by claimants. However, ATE insurance can be purchased by respondents as well, and the controversy over recoverability is less acute for the defendant side of the dispute. Moreover, there are simply no binding rules at all in international arbitration specifically addressing whether a successful party may recover its own ATE premiums from the losing side.

Regardless of the recoverability of the premiums, fee-shifting principles in international arbitration are far less predictable than in the English court system. Commonly used international arbitration rules, such as the 2012 ICC Arbitration Rules (art. 37), the LCIA Rules of Arbitration (art. 28), the ICDR/AAA International Arbitration Rules (art. 31) and the ICSID Arbitration Rules (art. 47), defer decisions about cost-shifting to the discretion of the arbitral tribunal, in recognition of the fact that national legal systems often differ significantly in how they handle the question. The uncertainty embedded inside the applicable arbitration rules is compounded by the diversity of cultural and legal backgrounds found on many arbitral tribunals. For example, arbitrators trained in the US legal system, where cost-shifting is highly unusual, may approach costs rulings from a different perspective than, say, Swiss practitioners trained in a civil law environment, where cost-shifting is routine.

The allocation of costs in the high-profile world of investment treaty arbitration is equally unpredictable. Prof. Susan Franck has analyzed a database of all known publicly available investment treaty awards through mid-2006,5 finding "seven different permutations" of decisions with respect to allocation between the disputing parties of the costs of arbitrators and institutions and the costs of attorney and other party fees and expenses. In those disputes, thirty three final awards (twelve in favour of claimants, nineteen in favour of respondents and two settlements) involved a "pay-as-you-go" approach towards both tribunal/institution and party costs, not cost-shifting. In fact, "pay-as-you-go" was the dominant result among the fifty two final awards examined in her study. Still, a significant number of awards produced a "loser pays" or "factor-dependent" result.6

Several implications arise from the uncertainty as to whether a cost-shifting approach or other approaches are likely to prevail in any particular international arbitration. Most obviously, this lack of predictability makes it harder for an arbitrating party to justify the expense of ATE insurance premiums in an environment (unlike the English courts) where the disputing parties cannot know in advance how legal fees and other expenses will be allocated between victor and vanquished.

Entirely apart from the core question of whether there is a "loser pays" approach that justifies the purchase of an ATE policy, prospective users of ATE insurance will wish to take into account the types of coverage limitations, exclusions, control of proceedings provisions and termination rights found in ATE policies. The wording of those contractual policy terms may differ, of course, from insurer to insurer and from policy to policy. So, the following generalizations are merely a guide to assist the arbitration community.7

An ATE insurance policy will cover all "opponent's costs". Consistent with the scope of recoverable costs in English courts, the sums available for coverage under these policies include "all costs, expenses and disbursements reasonably incurred" by the successful opponent. Thus, such "opponent's costs" are defined in one illustrative policy as follows:

5.12 Opponent's costs

"Opponent's costs mean all costs, expenses and disbursements that have been reasonably incurred by the opponent in the legal action. Where in the legal action orders are made both that costs be paid by the insured to the opponent and that costs be paid by the opponent to the insured, opponent's costs shall then be limited to the net sum (if any) payable by the insured to the opponent after all costs payable by the opponent to the insured have been set off, irrespective of whether or not payment is actually made.

No cover is provided in respect of any success fee to which the opponent's solicitor or barrister may be entitled."8

Notably, the success fees of opposing counsel are not covered by this definition. Naturally, the exclusion or coverage of success fees may be open for discussion with any particular insurer, especially if the forum is international arbitration where there are no rules barring a claimant's counsel from charging a success fee. But, as we all know, specially crafted insurance coverage is harder to come by and more expensive.

The insured respondent's own disbursements may also be covered, in addition to their opponent's costs. The same sample policy defines the insured respondent's covered "disbursements" in the following manner:

"5.5 Disbursements

5.5.1 Disbursements means expenses paid by the appointed representative [i.e., the person specified to the insurance company as the insured litigant's representative for purposes of making decisions about the dispute] to other parties that have been reasonably incurred on behalf of the insured in connection with the legal action (including a rateable proportion of the alternative dispute resolution fees that the insurer has agreed to in writing) which are not the subject of any agreement where expenses are paid depending on the outcome of the legal action, but not including:

(a) the appointed representative's costs;

(b) any VAT to the extent that the insured can recover such VAT from HM Revenue and Customs;

(c) any sum paid to a third party as a referral fee (whether or not described as such);

(d) any interest incurred as a result of the insured entering into an agreement to fund the legal action unless this has been agreed in writing in advance by the insurer by the issuing of an endorsement.

5.5.2 The premium will be deemed a disbursement for the purposes of clause 2.1.2. It is not taken into account in respect of the limit of indemnity."9

Interestingly, the interest incurred by a claimant in obtaining third-party funding is not a covered disbursement under this definition, absent specific approval from the ATE insurance company.

With that definition in mind, the same policy specifies its coverage of the litigant's disbursements, but again with limitations:

"… provided that:

(a) the court, or appellant court if the period of insurance is extended to cover an appeal, makes an order for costs (not being an interim order) against the insured; or the legal action is concluded with no order for costs against the insured and the insured has not been awarded damages or obtained the relief sought;

(b) the insurer shall only be liable for disbursements to the extent that the insured is not entitled to seek their recovery from the opponent or any other party;

(c) the insurer shall not be liable to pay any disbursements until the period of insurance finally ceases;

(d) the insurer shall provide no indemnity in respect of counsel's fees where counsel is acting under a conditional fee agreement.

In the event that the legal action is subject to appeal and at the end of the appeal's process the judgment of the court is amended to include an order for costs against the insured, or if an existing costs order against the insured is set aside, Section 2.1, above, will be construed as if the court had reached the same decision as the final appellant court."

According to this example, the insurer is only liable for the insured's disbursements to the extent that the insured is not entitled to seek their recovery from the opponent or any other party. Thus, the policy places collection risk on the litigant, not on the insurance company.

Risk-sharing principles are employed by ATE insurers just like they are employed in virtually all other insurance products, whether automotive, casualty, life or political risk insurance. Accordingly, an ATE insurance policy might not cover 100% of an opponent's costs and own disbursements. Rather, the policy may instead cover a specified percentage of the sum. To the same effect, ATE policies may have deductible floors and amount ceilings, with costs and expenses below the floor or above the ceiling remaining for the account of the insured arbitrating party.

Again, like other types of insurance products, an ATE insurance policy will typically exclude the insurance company from liability if the legal action involves fraud or:

• where the action is abandoned, discontinued, settled or lost as a result of dishonesty of the insured;

• if the covered items are incurred or increased as a result of a "failure on the part of the insured or the appointed representative to mitigate such liability";

• in claimant coverage, if the legal action is discontinued "as a result of the insured not having the funds to continue or not being willing to commit funds to continue the legal action";

• where the insured or its appointed representative "has failed to disclose material facts";

• where the insured's disbursements "in the insurer's opinion, have been incurred unreasonably or unnecessarily" or opponent's costs or own disbursements have been increased due to "any unreasonable delay or negligence or wilful act or omission by the insured or its appointed representative which in the insurer's opinion is prejudicial to the conduct of the legal action";

• unreasonable amendments to the insured's pleadings or failure to follow applicable rules or the court's directions;

• in the case of costs or disbursements associated with or an order of costs for security; or

• in the case of the opponent's insolvency.

This litany of exclusions from coverage ranges from fraud through unreasonable conduct of the proceedings to the insolvency of the opposing party. As these exclusions make clear, ATE insurance is certainly not an unconditional indemnity for costs or disbursements for the losing party.

2 "OUTCOME HEDGING" OR "LITIGATION BUYOUT" INSURANCE (OUTCOMES POLICIES)

Respondents care about far more than just protecting themselves against the costs and expenses of the dispute proceedings. It is of course the outcome that matters most. Unsurprisingly, respondents seek a way of mitigating the potential damages that may be assessed in the event they lose the arbitration. An outcomes policy is essentially a "stop-loss" contract "which caps the Defendant's potential liability at a certain level".10 While insurers will provide both financing and insurance products to cover outcomes and offer protection for both claimants and respondents, as a practical matter only the insurance product is useful for respondents. Outcomes policies may take the form of full buyouts of the risk of an adverse outcome, caps, hedges or other formulations - they are specially-crafted "bespoke" arrangements.

As is the case for other forms of insurance, the insurance policy will generally impose risk-sharing principles, establishing a "deductible" retained risk for the covered respondent before the insurance begins to provide protection and a policy ceiling above which the respondent is uncovered.

"The insured must retain an element of risk (i.e. an excess or retention) and the insurer will provide cover for any liability over this excess, up to a specified limit of liability."11

Full buyouts of the outcome risk by the insurance company may not have such deductibles, but one set of observers asserts that "Caps always have a deductible".12

Alternatively, insurance companies may offer a buyout of the risk in which the client sells and the insurance company purchases the risk of the arbitration outright. Additionally, for a fee, insurers may offer clients the ability to purchase an option today to compel the insurer to issue the outcomes policy in the future on pre-agreed financial and commercial terms. That option may help mitigate the risk of future arbitration, which has been threatened or can be seen on the horizon, but which has not yet crystallized. One major insurance provider describes an illustrative case involving a threatened claim by a minority shareholder against company management and directors.

"A private equity firm that owned a significant stake in a middle market size company issued letters that threatened litigation against this company in an effort to replace the company's current management and obtain control of the board of directors. The private equity firm alleged that the company's management and board breached various fiduciary duties resulting in a $63 million investment loss in the company. The company's Directors & Officers Liability Insurance policy denied coverage because of two separate exclusions. As a result of this potential litigation and the lack of available insurance, the company's independent directors threatened to resign their positions.

LBI [Litigation Buyout Insurance] provided the company with a two-year option to purchase a $5 million policy covering judgments and defense costs, with no retention. With the option in place, the independent directors agreed to remain on the company's board."13

Outcomes policies have been most heavily utilized in M&A transactions, where on-going litigation may serve as a barrier to successful completion of the transaction, for example by making it difficult to place a value on the target company/defendant and thus to agree on the acquisition price. The insurance policy serves to quantify the maximum risk of that litigation to the balance sheet of the target company, and thus permits the M&A transaction price to be agreed.

Outcomes policies can, of course, also be employed to cover a respondent's own risk in an on-going litigation to hedge against an adverse outcome, not just to help acquirer and target agree on an acquisition price in an M&A transaction. Public filings with the US Securities and Exchange Commission tell a useful tale of a transaction involving Samsonite, a public company in the United States. In 1998, Samsonite's board of directors tried to sell the company to investors in a "taking private" deal. But the deal fell apart, and Samsonite sought to borrow considerable sums to "releverage" itself. In the circumstances, Samsonite's public share price dropped by a significant amount. Unsurprisingly, disappointed shareholders promptly commenced class actions and derivative suits in various US courts, "alleging hundreds of millions of dollars of damages for a multitude of violations of federal and state securities laws in conjunction with the botched sale".14

That shareholder litigation, as a practical matter, blocked Samsonite from obtaining recapitalization funds in the financial markets. Samsonite reacted by purchasing litigation buyout insurance (LBI). With the LBI policy in place, Samsonite re-approached the financial markets and successfully obtained additional equity funds.

"An LBI policy permitted Samsonite to move forward. The insurer took over the litigation, assuming all potential liability. With the litigation risk transferred, Samsonite refocused on its recapitalization - this time without the baggage. Within six months, Samsonite raised $55 million of equity capital. It retired debt, and EBITDA was $91 million for 2000, up from $50 million in 1998."15

Astute readers will notice that, in the Samsonite story, the insurance company "took over the litigation, assuming all potential liability". Samsonite illustrates the situation of a "full buyout" policy, with no risk-sharing on the part of the client company, based on the insurance company's expectation about the ultimate cost of settling the litigation.

If ATE costs and disbursements insurance is expensive, then it will come as no surprise that buyout policies like the one in the Samsonite example and other outcomes policies can be far more expensive. For a buyout policy, as an example, the insurer will price the premium it charges relative to the amount it estimates that it will have to pay in settlement of the dispute.

"In a Buyout, in exchange for assuming the liability of a suit, the insurer looks to be paid a premium equal to what it sees as the expected settlement value, plus an additional amount for its trouble. In the Samsonite case, the insurer reportedly received a premium of $17 million."16

For that $17 million upfront premium, the insurance company accepted the risk that the settlement or court judgments for which it would be responsible might exceed its estimate. Of course, the insurance company also obtained the benefit of any excess of that premium over a final settlement or judgment, if the final payment came in below the insurer's expected settlement value. Accordingly, the role of experienced counsel in assessing settlement value for the dispute in arbitration, both on behalf of the insurer and on behalf of the client, is fundamental to the negotiation of the upfront premium payable by the client. Incidentally, that $17 million premium in 1998 is equivalent to about $23,770,434 in 2013, after giving effect to inflation at the U.S. Consumer Price Index over the intervening 15 years.

Premium pricing for outcomes policies, though, is generally determined on a case-by-case basis, with risk-sharing allocation part of the negotiations.

"Pricing, limits and size of retentions are determined on a case-by-case basis, depending on factors such as the underlying exposure, its severity and the program structure."17

Another case illustrates the cost of a prospective hedging transaction, rather than a buyout. A publishing company was simultaneously in the midst of private antitrust litigation with a challenged damages award of $25 million and the object of a $950 million acquisition. The company explored purchasing outcome insurance and received a quotation of a $2 million premium (about decade ago), although the litigation settled before the insurance was purchased.

"An example of a Hedge can be found in a private antitrust action between an industry leader in the publishing industry and its primary competitor. The industry leader found itself the defendant in a case that alleged it acted in an illegal manner by attempting to monopolize. A jury had awarded the plaintiff over $25 million. The defendant, however, successfully argued in a post-trial motion that the verdict was unsupported by the evidence, thereby winning a reversal. The trial judge's reversal was the subject of an appeal. While the appeal was pending, an investor purchased the company. The investor was reluctant to close the acquisition with this pending exposure on its target's books. As a means to address the prospective purchaser's reluctance, the company explored LBI. Had the case not been decided during negotiations, a policy insuring that the reversal would be upheld on appeal could have been purchased. The quoted premium was about $2 million. While expensive compared to traditional coverage, in the context of salvaging a $950 million acquisition, the cost was more than justified."18

Moreover, the point in time at which the respondent approaches the insurance company about an outcomes policy is important. "[I]f the litigation is advanced, insurers may consider the level of risk to be substantial, which may inflate the excess and premium or make it more difficult to secure commercially viable terms."19

Before issuing an outcomes policy, the insurance company will undertake substantial diligence regarding the risks it may be assuming and will engage its own counsel and other advisers for that purpose. Thus, the insurer's out-of-pocket costs in determining whether to provide the insurance protection can be substantial. The question whether the client or the insurance company will bear those costs, regardless of whether or not a policy is ultimately issued, will thus be the object of discussion between insurer and prospective insured before the application is considered.

The insured respondent will be required to make representations about the information provided to the insurance company as part of the diligence process to assess the claims and defences, as well as other significant matters. A respondent who, among other matters, has failed to disclose material information during the initial diligence process or to properly share information with the insurance company on an ongoing basis, contravened applicable law or engaged in fraud may find its outcomes policy subject to termination due to an event of default.

Moreover, diligence investigations and access to information by an insurance company for outcomes policies raises the same questions as third-party funding for claimants, which are discussed elsewhere in this volume, including the issue of whether such information-sharing results in waiver of any attorney-client or other privilege with respect to the information shared.

The extent to which the insurance company has the right to control the arbitration defence is also potentially controversial. Even more so than in the case of third-party funding for claimants, providers of outcomes hedges, caps or buyouts to respondents may be concerned about control rights with respect to the arbitration. Outcomes policies may consequently grant the insurance company a significant measure of control over choice of defence counsel, the course of prosecution of the defence and settlement negotiations, since the insurance company is bearing a substantial part of the risk of the outcome.

"An insured also needs to become comfortable with the notion of ceding, or, at a minimum, sharing, control of the litigation with the insurer. An insurer will be quite concerned over major strategic decisions, choice of counsel, settlements, etc. For some insureds, the idea of sharing - let alone ceding - control over litigation does not come easily. Quite naturally, a defendant wants to maintain exclusive control over its own destiny. It also is quite reasonable for the insurer to want to have the ability to protect its position and utilize its expertise in handling litigation (which often is extensive) to minimize the exposure.

In a Buyout, the insurer assumes 100 percent of the liability and covers defense costs. Thus, it typically requires that complete control over the handling of the litigation be ceded to it. This may include a 'hammer clause' allowing the insurer to settle the case. Of course, the insured company is required to continue to cooperate and participate in the subject litigation. After all, it still is the defendant. An insured should not find this difficult to swallow because there is little downside to the insured. In some cases, insureds have been able to negotiate specific rights regarding the insurer's ability to settle, particularly for nonmonetary damages. However, in a Cap, and in certain Hedges, both the insured and the insurer are at risk. Hence, each may have very specific, and perhaps opposing, views of how to handle the litigation. In the Cap described above, the insurer was granted exclusive control over the handling of the case. Settlements required the insured's consent. The opposite could also apply in a Cap, the insured maintaining control over the litigation and the insurer having veto power on major decisions and settlements."20

In fact, it is an overstatement to say that an insured company selling the risk in a buyout "should not find [ceding control of the arbitration and settlement] difficult to swallow because there is little downside to the insured". Even if the insurance provider acquires all direct financial risk by means of a buyout, the insured client may still face significant reputational risk. Additionally, its shareholders, officers, board members and affiliates may face follow-on consequences from the resolution of a particular dispute. These consequences may be of a financial, reputational and sometimes even criminal nature. The issue of control rights can therefore be particularly sensitive for some respondents. For state respondents in investor-state arbitrations (whether in international commercial arbitration or in investment treaty arbitration), the question can be acute, since the state stands as political fiduciary for its citizens.

The confidentiality of the outcomes policy is another important issue. For example, mere knowledge of the existence of such a policy may embolden claimants to pursue the claim and focus on the maximum policy ceiling as a realistic settlement outcome. Consequently, an outcomes policy will invariably contain a confidentiality clause requiring the consent of the insurance company before even its very existence can be disclosed.

There may be practical reasons why a respondent nevertheless wishes to disclose the existence of an outcomes policy, for example to reassure shareholders and creditors that a contingent litigation liability is no longer a serious risk to the company's health and prospects. "Samsonite, Oxford Health Plans, Inc., and … Synopsys, Inc., for example, issued press releases announcing their use of LBI."21

At times, moreover, disclosure of the outcomes protection is mandated by applicable law. In that situation, the insurance company must consent to such disclosure if it wishes to sell the product at all to the particular client. Illustratively, the purchaser of the policy may be a public company with securities that are trading on US securities exchanges. If the insurance is "material" to the company, then US securities law will compel disclosure of the existence of the hedge and its key terms. Corporate respondents seeking to obtain outcomes policies are well-advised to obtain the views of reputable counsel about mandatory disclosure in each of the relevant jurisdictions in which their securities are traded. Securities counsel can advise not only on whether disclosure is required but also as to the extent of disclosure - must some or all of factors such as policy size, identity of the issuer, cost of premiums, control rights and other material terms, or indeed the actual policy itself, be publicly disclosed under applicable securities law?

Companies in regulated industries, such as financial institutions, will also be aware of the need to be able to disclose such insurance policies to their regulator, occasionally in a number of jurisdictions.

State entities may similarly find that the existence and terms of an outcomes policy must be publicly accessible in compliance with sunshine and freedom of information statutes.

In addition to public disclosures, respondents will wish to consider whether the existence of the outcomes policy and its terms may be subject to document production and other forms of information exchange within the international arbitration in question. Even if the popular International Bar Association (IBA) 2010 Rules on the Taking of Evidence in International Arbitration are employed in the arbitration, they do not shed much light on this question. The principles for document production under article 3 of the Rules are clear, but of course their application will depend on the particular facts and circumstances of the dispute, as well as the predilections of the particular arbitrators.

"Article 3: Documents

…

3. A Request to Produce shall contain:

(i) a description of each requested Document sufficient to identify it, or

(ii) a description in sufficient detail (including subject matter) of a narrow and specific requested category of Documents that are reasonably believed to exist…(b) a statement as to how the Documents requested are relevant to the case and material to its outcome; and

(c) (i) a statement that the Documents requested are not in the possession, custody or control of the requesting Party or a statement of the reasons why it would be unreasonably burdensome for the requesting Party to produce such Documents, and

(ii) a statement of the reasons why the requesting Party assumes the Documents requested are in the possession, custody or control of another Party."

Moreover, a request for production of an outcomes policy in the arbitration (and even a cross-examination question about the existence or non-existence of such a policy) will likely be met with an objection under article 9.2 of the IBA Rules. Such objections are unexplored waters for disclosure of outcome insurance just as they are for the disclosure of third-party funding of claims. Article 9.2 permits an arbitral tribunal to exclude or limit evidence or document production on various grounds, including lack of relevance or materiality, legal impediment or privilege, compelling grounds of confidentiality, compelling grounds of special political or institutional sensitivity (including evidence that has been classified as secret by a government) and considerations of fairness.

"Article 9: Admissibility and Assessment of Evidence

…

2. The Arbitral Tribunal shall, at the request of a Party or on its own motion, exclude from evidence or production any Document, statement, oral testimony or inspection for any of the following reasons:

(a) lack of sufficient relevance to the case or materiality to its outcome;

(b) legal impediment or privilege under the legal or ethical rules determined by the Arbitral Tribunal to be applicable;

(c) unreasonable burden to produce the requested evidence;

(d) loss or destruction of the Document that has been shown with reasonable likelihood to have occurred;

(e) grounds of commercial or technical confidentiality that the Arbitral Tribunal determines to be compelling;

(f) grounds of special political or institutional sensitivity (including evidence that has been classified as secret by a government or a public international institution) that the Arbitral Tribunal determines to be compelling; or

(g) considerations of procedural economy, proportionality, fairness or equality of the Parties that the Arbitral Tribunal determines to be compelling.

3. In considering issues of legal impediment or privilege under Article 9.2(b), and insofar as permitted by any mandatory legal or ethical rules that are determined by it to be applicable, the Arbitral Tribunal may take into account:

(a) any need to protect the confidentiality of a Document created or statement or oral communication made in connection with and for the purpose of providing or obtaining legal advice;

(b) any need to protect the confidentiality of a Document created or statement or oral communication made in connection with and for the purpose of settlement negotiations;

(c) the expectations of the Parties and their advisors at the time the legal impediment or privilege is said to have arisen;

(d) any possible waiver of any applicable legal impediment or privilege by virtue of consent, earlier disclosure, affirmative use of the Document, statement, oral communication or advice contained therein, or otherwise; and

(e) the need to maintain fairness and equality as between the Parties, particularly if they are subject to different legal or ethical rules.

4. The Arbitral Tribunal may, where appropriate, make necessary arrangements to permit evidence to be presented or considered subject to suitable confidentiality protection."

If there are precedents for how arbitral tribunals have resolved demands by claimants that respondents produce their outcomes policies in international arbitration, those precedents are themselves currently locked away behind bars of confidentiality.

CONCLUSIONS

The premiums charged for ATE insurance and specially negotiated "bespoke" outcomes policies limit the attractiveness of these third-party risk management products for respondents. The unpredictability of cost-shifting in international arbitration is another important deterrent to the use of ATE insurance. And the availability of outcomes policies is heavily dependent upon the particular risks to be assumed by the insurance company. Even though these barriers are high, responsible counsel will discuss with thoughtful clients whether or not these products are sufficiently valuable to investigate in the proper circumstances. But respondent and counsel alike should beware - these seas are marked "hic sunt dracones" (Here be Dragons) on the arbitration map. Navigate carefully.

1
Pinsent Masons, Advice Note: CFAs and ATE Implications for you and your opponent (2009), available at: http://www.out-law.com/en/topics/dispute-resolution-and-litigation/litigation-funding/cfas-and-ate-implications-for-you-and-your-opponent (last visited March 1, 2013). This report predates the 2013 changes in the ATE insurance product mandated by the Legal Aid, Sentencing and Punishment of Offenders Bill (in response to the Jackson Report) discussed below, which may lead to a market increase in ATE insurance premiums.

2
Jackson, Review of Civil Litigation Costs: Final Report, December 2009 (TSO 2010), at para. 2.2 and ch. 9.

3
Id., a tpara. 1.3. Inf ac t, English litigation fundersro utinely require a claimant non-recourse borrower to also obtain ATE insurance as security for costs in the event of an unsuccessful outcome.

4
Royal Assent, 2 May 2012.

5
Franck, 'Rationalizing Costs in Investment Treaty Arbitration', Wash. U.L. Rev. 88 (2011) p. 769.

6
Id., a tpp .809-811.

7
For illustrative purposes, I refer in this paper to a specimen QBE Insurance (Europe) Limited ATE Policy, available at: http://www.qbeeurope.com/documents/casualty/ specialist_liability/QBE_ATE_Wording.pdf (last visited March 1, 2013).

8
Id.

9
Id.

10
Turner, 'Options for Transferring Litigation Risk', Journal of the British Insurance Law Association 123 (2011) pp. 65, 70.

11
Id.

12
Kerxton and Blitz, 'Litigation Buyout Insurance: Removing Pending Legal Actions from the Deal', LTC Matters: Legal Update for Long Term Care and Assisted Living Providers (Mintz Levin Cohn Ferris Glovsky & Popeo PC, March 2003) at p. 3.

13
Chartis Mergers & Acquisitions Group, Litigation Buyout Insurance, at p. 2, removed from Chartis website but available from the Internet Archive Wayback Machine at: http://web.archive.org/web/20120619001254/ http://www.chartisinsurance.com/us-litigation-buyout-insurance_295_182176.html .

14
Kerxton and Blitz, supra n. 12, at p. 2.

15
Id.

16
Id.

17
Chartis Mergers & Acquisitions Group, supra n. 13, at p. 1.

18
Kerxton and Blitz, supra n. 12, at pp. 2-3.

19
Turner, supra n. 10, at p. 70.

20
Kerxton and Blitz, supra n. 12, at p. 3.

21
Id., at pp. 3-4.