Palladian Partners LP & Ors v The Republic of Argentina & Anor

The claim is concerned with amounts said to be due to each of the Claimants under certain Euro denominated securities, originally issued by the First Defendant, the Republic of Argentina (“the Republic”) in 2005 and 2010 and linked to its Gross Domestic Product (“GDP”) (“the Securities”).

The Republic issued the Securities as part of a restructuring which gave creditors 25 to 35% of what they were originally owed. The Securities provide for payment each year of an amount linked to GDP (“the Payment Amount”), if certain pre-conditions are satisfied.

One such pre-condition to payment under the Securities in any given year (“the Performance Condition”), relates to the year on year economic performance of Argentina. For any given year, it compares percentage growth in Actual Real GDP measured in constant (base year) prices to percentage growth in a prescribed Base Case GDP also measured in base year prices. Initially and until 2013 the base year used (“the Year of Base Prices”) was 1993.

The Securities give the Republic the freedom to rebase its GDP statistics to a different Year of Base Prices. In 2013 it did so, and as a result, the Ministry of Economy announced the Payment Condition for that year was not met and that no payment was due to any of the Holders of the Securities.

The Claimants take issue with that decision and process. They say that they are owed the amount which would have been payable if the Year of Base Prices had not been changed, an amount totalling between allegedly €525-€645 million (depending on the basis of calculation). They have claimed in this court for a variety of forms of relief. The claim is for breach of contract, but also comprehends declaration, order for payment, damages or specific performance.

By this Application, issued on 4 March 2020, the Republic seeks to short circuit this claim by seeking two orders, which if granted would dispose of the claim. Those two applications are:

Unsurprisingly that application is hotly contested by the Claimants. In a nutshell their position is that:

The legal principles underpinning the exercise of the powers to grant summary judgment and to strike out were not in issue, and need not be considered here.

The background to the Securities involves the fact that the original Holders were formerly owed money by the Republic and as part of the arrangement for the issue of the Securities took what might be colloquially described as a considerable haircut. The Securities thus at their inception provided billions of dollars of debt relief to the Republic. At the same time they provided for contingent additional payments to the holders of the Securities - if but only if certain thresholds were exceeded.

Under the Securities, which have a 30-year lifespan from 2005-2034, the question of such further payment is assessed annually by reference to calendar years (defined as “Reference Years”).

The question of whether a payment is to be made is determined by the satisfaction of three Conditions – being (i) the GDP Level Condition, (ii) the Performance Condition and (iii) an Aggregate Payments Condition:

“Notwithstanding anything to the contrary hereunder, Holders of this Security shall not be entitled to receive any payment pursuant to this Security in respect of any Reference Year unless (i) Actual Real GDP for such Reference Year is greater than Base Case GDP for such Reference Year [the GDP Level Condition], (ii) Actual Real GDP Growth for such Reference Year is greater than Base Case GDP Growth for such Reference Year [the Performance Condition], and (iii) the aggregate amount of all payments made by the Republic hereunder, when added to the amount of such payment, does not exceed the Payment Cap [the Aggregate Payments Condition].”

The determination of any Payment Amount, is contingent on satisfaction of those Conditions. Determination of the Payment Amount involves a formula to determine what is defined as “Excess

The Payment Amount is defined thus, (with one of the so called Binding Effect provisions in dispute here forming the last part of the clause):

““Payment Amount” means, for any Payment Date, an amount equal to (i) the Available Excess GDP (converted to [U.S. dollars] [euro] [Other currency]) for the Reference Year corresponding to such Payment Date, multiplied by (ii) the notional amount of this Security outstanding as of such Payment Date; provided that, if for any Payment Date, the Payment Amount determined in accordance with the foregoing would, when added to all prior Payment Amounts paid by the Republic hereunder, exceed the Payment Cap, the Payment Amount for such Payment Date shall instead be an amount equal to the Payment Cap minus the sum of all such prior Payment Amounts. The Payment Amount shall be determined by the Ministry of Economy on the Calculation Date preceding the relevant Payment Date. All calculations made by the Ministry of Economy hereunder shall be binding on [all relevant persons including holders], absent bad faith, willful misconduct or manifest error on the part of the Ministry of Economy”.

Whilst the Payment Amount produced by the formula may be positive in any given year, the Republic is only required to actually make a payment in years when each of the Conditions are also met. So if the Performance Condition is not met, the Republic will not go on to calculate the Payment Amount, even if that separate calculation would produce a positive figure. GDP is therefore critical to the obligation to make a payment. Page 1 of the Securities makes this very clear: “The only amounts payable in respect of this security are the payments contingent upon and determined on the basis of the performance of the Gross Domestic Product of The Republic of Argentina (“The Republic”) referred to herein”.

The Securities are governed by a document called the Trust Indenture, the terms of which were incorporated into the Securities; a point which becomes relevant to the arguments on the Binding Effect provisions at the heart of this dispute.

At the heart of the dispute is the operation and application of the “Performance Condition”. As noted above, for the Performance Condition to be met in respect of any Reference Year, “Actual Real GDP Growth” must exceed “Base Case GDP Growth”.

As for the relevant terms;

Base Case GDP is then set out in a chart on page G-7 of the Securities. The chart gives the Base Case GDP in millions of “constant 1993 pesos” for each year – in other words an actual figure for GDP in 1993 prices. That chart on its face envisages a steady growth in GDP from 287,000 million to 674,000 million in that period.

The Republic relies heavily on the fact that if one applies the contractual definition of Base Case GDP Growth to the figures derived from the table, Base Case GDP Growth is a readily ascertainable rate that is slightly above 3% in the early years in the life of the Securities (2005-2014) and evens out to exactly 3% from 2015 to 2034. For 2013, Base Case GDP Growth extracted from the figures was 3.22%. The Republic says that it is part of the relevant factual background to the Securities that these figures were “baked in” to the terms and that it follows that this (and no other) level of growth is part of what the holders of the Securities contracted for.

The Performance Condition thus assesses whether the percentage year-on-year growth in Actual Real GDP for the given Reference Year is higher or lower than the percentage year-on-year growth in the Base Case scenario. To meet the Performance Condition, the Actual Real growth rate must be higher than the Base Case growth rate. The question which arises is how that operates when the Year of Base Prices is changed.

Mr O’Rourke’s evidence, which was not challenged on this point, was that there is no single way or absolute way to measure GDP – but it must always be by reference to a specified year of base prices. Further GDP for 2020 measured in constant 2005 prices will be different from GDP for 2020 measured in constant 2015 prices. This is not just about inflation or the value of money; it reflects changes to the categorisation and weighting of the various goods and services which are used to determine GDP.

A change to the year of base prices can thus make a very significant difference to the assessment of Actual Real GDP and Actual Real GDP growth because of the differences in weightings and methodologies. Further, as was apparent from the evidence, the relationship between two different years of base prices is not linear.

The Year of Base Prices to be used for the purposes of the

The existence of a defined Year of Base Prices did not preclude the Republic from deciding to publish figures in a different Year of Base Prices if, during the 30-year lifetime of the Securities it decided to publish Actual Real GDP figures in a different Year of Base Prices. Where it did so, that new base year would become the “Year of Base Prices”. This process is commonly referred to as a “rebasing”.

However, the Claimants submit that in that situation the provision referred to by the parties as the “Adjustment Provision” must be applied. That provision is not actually a separate clause, but forms part of the definition of Base Case GDP. It states, following the table setting out the base case GDPs for each year:

“…Provided that if the Year of Base Prices employed by INDEC for determining Actual Real GDP shall at any time be a calendar year other than the year 1993, then the Base Case GDP for each Reference Year shall be adjusted to reflect any such change in the Year of Base prices by multiplying the Base Case GDP for such Reference Year (as set forth in the chart above) by a fraction, the numerator of which shall be the Actual Real GDP for such Reference Year measured in constant prices of the Year of Base Prices, and the denominator of which shall be the Actual Real GDP for such Reference Year as measured in constant 1993 prices.”

For present purposes I need not go into exactly how this provision is said to work. It is enough to say that the Claimants contend that following the approach which they say is correct, the result, in terms of the correct figure for the Performance Condition, would be the same as if 1993 had remained the year of Base Prices. In other words a change to the Year of Base Prices does not affect the amount payable to the Claimants. And by extension it is the Claimants’ case that it is not possible for the Claimants to be deprived of the amount they would have received if there had been no change to the Year of Base Prices by a change to the year used for the calculation of GDP. As the Claimants put it: “the goalposts cannot be moved. … It guards both against the moral hazard of a change designed to reduce measured GDP and so avoid paying, and against incidental effects of a rebasing”.

The Republic’s primary position is that the Adjustment Provision has no application to Base Case GDP Growth (which is “baked in”) and hence to the Performance Condition. It follows that there is no change to Base Case GDP Growth following a change in the Year of Base Prices. It also submits that the Adjustment Provision does not apply if, as was the case here, figures based on 1993 Base Case cease to be published, and that to the extent necessary if it does apply, the correct application of the provision reaches the same conclusion, because one uses a single adjustment factor, represented thus: Base

I should add something briefly about the role of INDEC. INDEC, the Argentinian National Institute of Statistics and Censuses, is the Argentinian governmental agency established by statute, responsible

The background to the Republic’s actions is detailed, and is clearly capable of being contentious. In brief summary the position appears to be as follows.

It would seem that there is a consensus that the base for assessing a GDP should be reviewed and updated because measurements may fall out of line with reality. So in its Defence, the Republic says that the reason why INDEC discontinued measuring real GDP in 1993 prices and started using 2004 prices was that measuring real GDP in 1993 prices no longer accurately reflected the Republic’s economy and the IMF had urged the Republic, during 2012 and 2013, to revise its GDP data with an updated base year.

The process which led INDEC to change the Year of Base Prices in March 2014 had begun some years earlier. I have been shown one 2007 document showing that already at that stage the Republic was looking to the development of a new system built with 2004 as the base year. There is also reportage which suggests that some commentators considered that the Republic’s GDP figures were overstating growth.

This chimes with INDEC’s Activity Level Progress Report dated 27 March 2014, which initiated the reporting of GDP data in 2004 prices, and referred to “a large and accurate work of more than three years during which a deep analysis was made of a large quantity of statistical information from different sources …”.

An IMF Press Release dated 1 February 2012 records the decision of the IMF’s Executive Board calling on Argentina to implement special measures, within 180 days, to address the quality of reported GDP data, with a view to bringing the quality of the data into compliance with the obligation under the IMF’s Articles of Agreement. Another IMF Press Release dated 18 September 2012 records the decision of the IMF’s Executive Board to require a report on Argentina’s response to the Fund’s concerns by 17 December 2012.

An IMF Press Release dated 1 February 2013 speaks of the issuing of a declaration of censure against Argentina and calling on Argentina to adopt remedial measures to address the inaccuracy in GDP data without delay, and in any event by no later than 29 September 2013. The measures required were aimed at aligning GDP with international statistical understandings and guidelines that ensure accurate measurement.

The IMF Managing Director’s Report on Argentina dated 13 November 2013, outlined remedial steps being undertaken by Argentina to address the improvements required to GDP reporting, including to “update, as a matter of priority, the current 1993 base year of the national accounts using information from the 2004 Economic Census and other comprehensive sources. According to international guidelines, the base year in national accounts should be updated as appropriate to reflect the pace of structural change in the economy.” That work was required by September 2014.

It further notes: “Since the February 2013 Board meeting, the authorities have shared with staff an updated work schedule for improving GDP data. The updated plan envisages that a revised GDP series from 2004 with an updated base year will be published on the INDEC website by March 2014.”

The IMF’s Decision dated 9 December 2013 recorded Argentina’s intention to address shortcomings in its GDP data and noted that, by 31 March 2014, Argentina would release to the public revised GDP data with an updated base year for Q4 2013, and as far back as feasible. As the Decision also recorded: “Argentina’s failure to adopt any one of the specified actions called for by the Executive Board by the relevant deadline may result in the issuance of a complaint and the adoption of a declaration of ineligibility to use the general resources of the Fund …”.

This background overlaps with the developing economic picture:

In March 2014, INDEC discontinued Actual Real GDP in 1993 prices and thereafter adopted Actual Real GDP in 2004 prices, which was a rebased calculation using a different model with different weightings to 1993 prices. On this basis, the Minister of Economy announced that Actual Real GDP Growth for 2013, calculated in 2004 prices, was

The last publication of Actual Real GDP in 1993 prices thus occurred in December 2013, for Q3 2013. For Q4 2013 and for the full year 2013, INDEC published Actual Real GDP in 2004 prices, not in 1993 prices. Final GDP figures for Q4 2013 and FY 2013 in 1993 prices were not published.

The Republic announced in its Press Release of 12 December 2014 that:

“… it is not required to make payments during 2014 in connection with the GDP-linked Securities issued in 2005 and 2010 debt exchanges. The annual growth in Actual Real GDP (as such term is defined in terms and conditions of GDP-linked Securities) for 2013 did not exceed the growth rate in Base Case GDP (as such term is defined …) for that year; therefore, one of the conditions required for payment under the terms of the GDP-linked Securities was not met.”

The Republic says that it follows that as a consequence of the change in Year of Base Prices, applying the language of the relevant definitions used in the Terms and Conditions to the situation in the present case:

It is the Republic’s contention in its Defence that, notwithstanding the change in Year of Base Prices, Base Case GDP Growth between Reference Years 2012 and 2013 was not to be calculated using adjusted Base Case GDP as Claimants contend, and that the “Adjustment Provision,” which, as construed by the Claimants, relies on a formula that requires an input that was and is unavailable, is not applicable.

The Claimants disagree, contending that the Performance Condition would have been met if the Ministry of Economy had used adjusted Base Case GDP Growth in 2004 prices in 2013.

These contentions are, however, not issues which the Court needs to resolve (or will ever need to resolve) if the Republic’s applications are well founded: if, in other words, on a proper interpretation of the Terms and Conditions, the Ministry of Economy’s determination that no Payment Amount was due to the Claimants is binding on the Claimants by reason of the Binding Effect provisions contained in the Terms and Conditions.

It is the Republic’s case that the Terms and Conditions contain two Binding Effect provisions. The key provision is found in the definition of “Payment Amount”:

“All calculations made by the Ministry of Economy hereunder shall be binding on … all Holders absent bad faith, willful misconduct or manifest error on the part of the Ministry of Economy”.

In addition, the definition of “Excess GDP” provides:

The Construction Issue concerns the proper construction of the Binding Effect provisions. Put shortly, the Construction Issue is whether the Binding Effect provisions apply (as the Republic contends) to all calculations made by the Ministry of Economy in respect of the Payment Amount, including, in particular, as to whether the Performance Condition is met; or whether (as the Claimants contend) those provisions apply only to the calculations found in the definitions of “Payment Amount” and “Excess GDP”, and do not apply to the determination of whether the Performance Condition has been met.

The submissions of the Republic on this point were that the position is in summary:

The Republic also submits that this approach fits with the overall contractual scheme and reflects a plausible case on commercial purpose. On this it is said that there is a need for a comprehensive Binding Effect provision to promote certainty, against a background where none of the holders had any involvement with the negotiation of the Securities, and many of them would become Holders only long after the securities were issued and where the ascertainment of the Payment Amount, if any, involved a number of calculations, giving scope for a multiplicity of challenges, absent such a provision. Drawing the line in a place where no Holder is able to challenge any of the many calculations, save in cases where bad faith, wilful misconduct or manifest error is proved, makes perfect sense.

The Republic submits that this issue of construction is one which is entirely suitable to be determined summarily, and that I should grasp the nettle and do so, in its favour.

The Claimants dispute the Republic’s construction of the Binding Effect provisions, contending that:

As is now well understood, the exercise of construction is an iterative process. I start with a simple reading of the relevant provision. The first indication is not one which favours the Republic. That is because the so-called “Binding Effect Provisions” are not in truth free-standing provisions, but tail ends to the definitions of “Payment Amount” and “Excess GDP”. That location on its face is suggestive that what is intended is to cover the calculations described within those provisions.

That impression is to some extent reinforced by the fact that Excess GDP is but one of the defined versions of GDP and (ii) it is the only form of GDP referred to within the Payment Amount provision. The appearance therefore, is of a “ring-fencing” of binding effect, to cover the determination of the Payment Amount, and the specific form of GDP which feeds into that calculation, and no more.

I should add that I was not persuaded that the Claimants’ submissions on the use of binding effect language in the Excess GDP provision shed much light on the matter. It might be the case that two such provisions slightly negatives the suggestion that “hereunder” was to be read widely; but the Republic’s point as to belt and braces drafting, by reference to Metlife v JP Morgan offers a very cogent possible explanation.

One then pans out to the consideration of the provisions in the wider context. There are a number of aspects to this. The analytical arguments so clearly and attractively deployed by Mr Valentin QC for the Republic certainly form part of this. They have a logic. They go so far as to persuade me that the initial reading is not clear beyond peradventure.

In the end however I have preferred the contrary textual/contextual arguments, even without resort to the question of whether the “Binding Effect Provisions” should be regarded as exclusions, and thus subject to a strict approach to construction.

Against the background of the pure drafting points, the Claimants’ arguments as to the distinction which can be drawn in the wider wording are the more powerful. There is a real distinction, reflected in the wording of the Securities, between satisfaction of the Conditions, and calculation of the Payment Amount. The right to payment is in Clause 2(a). It is expressly stated to be “subject to the conditions set out in Paragraph 2(b)”. Clause 2(b) is then presented separately. It is cast in terms of entitlement and conditionality, not calculation. It does not read as a provision which provides for calculations to be done. Indeed it was the Republic’s own case that it was able to declare that (to put it neutrally) no payment would be made without performing a calculation, because it was able to see simply that one of the Conditions was not met. In essence, it had its “baked in” 3.22% and INDEC published GDP figures that produced a growth figure of 2.93% for 2013. No calculation was necessary.

This distinction was in fact echoed by the Republic in the way in which it presented its decision. Its Press Release stated: “it is not required to make payments ... The annual growth in Actual Real GDP … for 2013 did not exceed the growth rate in Base Case GDP …; therefore,

That difficulty in presenting the process here under the spotlight as one simply of calculation was also apparent, despite careful consideration and acute sensitivity to the issue, in the way the submissions before me were phrased. So Mr Valentin spoke of an exercise: “both to decide whether a payment amount is due, …but also in determining what if any payment amount is due”. At another point: “The calculations which are challenged by the Claimants’ claim all ultimately go to the question of whether a payment amount was due and, if so, in what amount." Or in closing: “…the question [is]: is a payment due and, if so, in what amount?” What this illustrates is that there is no way of integrating the two aspects sensibly.

That fundamental disjunction was illustrated by Ms Prevezer QC in her submissions, via an analogy constructed by Mr Barden. Arsenal Fan A promises Arsenal Fan B that should their team win the FA Cup he will pay B £1 for every fan who attended the match, with A’s count of fans being final and binding. That bet comprises a condition – do Arsenal win, which is binary, and a separate calculation, which is the subject of the final and binding determination. Our situation is slightly more complicated in that there are three Conditions, not one and one or more of them may involve some calculations. But the question of satisfaction is binary, and the calculations to quantify the outcome of the Securities are different and separate from those which have to be performed to calculate the Payment Amount.

That sense of distinction is also reinforced by the fact that the “Binding Effect Provisions” are specifically linked to calculations to be performed by and errors etc on the part of the Ministry of Economy; however, in relation to various of the processes involved in the determination of satisfaction of the Conditions, it is entirely unclear as to whether they are to be performed by the Ministry of the Economy. Indeed as is apparent from the earlier points as regards the actual determination of the Performance Condition it appears that such calculations as there are, are (i) performed by the drafters of the Securities and (ii) performed by INDEC in calculating GDP.

Here there is an overlap with the subject matter of the main issue in the case, as to the Adjustment Provision. The non-satisfaction of a condition may be caused by the switch from 1993 prices to 2004 prices, and/or the non-application of the Adjustment Provision and/or the non-publication of parallel data in 1993 prices. But it is by no means certain that any of these are decisions of the Ministry of Economy. That makes it uncomfortable to conclude that the “Binding Effect Provisions” do apply to the question of satisfaction of the Conditions, which is entirely dependent on such exercises.

As for the suggestion that this argument is undermined by the fact that no distinction is drawn under the Terms and Conditions between “mechanical calculations” to be carried out by the Ministry of Economy on the one hand, and the other calculations, all of which are an integral part of the process of determining the Payment Amount, if any, on the other; that is in effect a “bootstraps” argument. Nor do I consider that, as the Republic sought to suggest, such a distinction would be impossible to maintain. Although the exercises of determining satisfaction of the Conditions and calculation of the Payment Amount are to some extent inter-connected, the Payment Amount provision and the inclusion of the “Binding Effect Provisions” within it makes clear which calculations are covered – those explicitly described in that provision.

Against this background I do consider that the Republic’s argument places an excessive burden on the word “hereunder”. As a matter of language one would not tend to expand the word out from the specific clauses where it is used, and certainly not to the extent which the Republic seeks to do here. It was probably for this reason that the argument based on incorporation of the Trust Indenture and a very wide definition of hereunder was originally advanced. That argument was rightly taken rather lightly in the oral submissions. It is an overstrained analysis. While technically the Terms and Conditions of the Securities incorporate the definitions of the Indenture, they have their own definitions section at Clause 1, in which “hereunder” does not feature as a defined term. There is nothing to suggest that the draftsman had the second-hand defined meaning in mind.

Some further slight reinforcement to the preliminary conclusion on the words is also obtained from noting the slight oddity of putting a Binding Effect provision of general effect as an embedded feature of a paragraph of a specific definition. While I entirely see Mr Valentin’s submission that the Binding Effect provision is in the right place, one still might expect sophisticated legally-advised parties intending it to be read as a general provision to have included it as a separate provision. This is the more so when, on looking at the wider terms, an examination indicates that these parties did just that on other occasions – there are more general limitations of liability in a number of places, such as Clause 5.2 of the main Indenture, sub-paragraphs (vi), (vii), (xxi) and (xxviii).

Nor in the end is there any difficulty about the fact that the Binding Effect provisions make reference to “bad faith” and “wilful misconduct”. This argument is based on a false dichotomy introduced by the terms in which the argument was phrased between assessments and “mechanical calculations”. The process of calculating the Payment Amount, once the Conditions have been fulfilled, does involve processes which involve an element of subjective judgement; for example if an official deliberately, in performing the Payment Amount calculation, got the nominal amount of the securities or total amount of prior payments wrong. Further one must always bear in mind the tendency to over-inclusive drafting. The scope for wilful misconduct may be very small, but it is there (and as the examples just given illustrate, capable of being highly significant). That being the case, the use of familiar boilerplate phrases to cover all contingencies is hardly surprising.

As for the argument on the definition of “reserved matters” in the context of modifications, I was not persuaded that this advanced matters at all. While it is right that the question of construction has to be performed in the wider context of the contract as a whole, the modifications regime is a separate area and unlikely to offer much robust guidance (bearing in mind the surplusages and inconsistencies which often creep in in drafting complex documentation). The Republic suggested that the definitions implied that the Claimants were wrong, because if it were not so calculation would be a reserved matter, but the Conditions would not, which would be an oddity. Aside from the point made above, as to inconsistency/surplusage, there is also the point noted by the Claimants, that such a distinction might be precisely because of the existence of the Binding Effect provisions.

Having reached this point by way of the normal exercise of construction, it is not necessary to decide whether such a provision would fall to be treated as an exclusion, or, as the Republic submitted, simply as a term which defines the ambit of the inclusions, by analogy with Impact FundingSolutions v Barrington Services [2016] UKSC 57 [2017] AC 73. Had the point required to be decided I would have formed the view that it was the former.

The point was made by the Republic, by reference to Impact, that just because words of exception are used does not make a clause an exemption clause. As noted at p. 86 of that case: “Words of exception may be simply a way of delineating the scope of the primary obligation.”

However in this case the words which are used are, as Ms Prevezer submitted, classical words of exception. Nor is there a factual parallel with Impact which might give cause for reconsideration. Impact was a case where the insurance was for legal indemnity insurance covering liabilities arising out of legal services. The clause in question defined areas of potential liability which would not be covered (such as provision of goods and services in the course of legal services), or provision of finance or other benefits in association with the provision of legal services. In effect the insurer had carved out by an objective description in the contract, certain liabilities over which questions might arise as not being insured.

So in either event, whether or not the clause is to be regarded as an exclusion clause, the same point is reached at this point in the exercise of construction.

That effectively leaves the question of commercial purpose. This is itself problematic for the Republic’s application because a construction argument which hinged on commercial purpose for success would rarely be suitable for summary determination. Here it most certainly would not. The Republic certainly makes a coherent case as to commercial purpose. That is clearly a commercial purpose which could well operate. But it could not be said, at this stage and without further evidence, that it was unarguably the commercial purpose which underpinned this drafting. Ms Prevezer certainly took issue robustly with this suggestion:

“it is actually slightly patronising to the sort of sophisticated people who invest in these securities. They well understand the way in which different growth rates and years of base prices work and the securities opted for an approach which adjusts each year to produces the fairest and most accurate result rather than a simplistic approach that Mr Valentin contends for ..

it is particularly ironic to advance a case based on everyone knowing where they stand in a situation where the Republic has rebased mid-year with the effect it says of moving the goalposts. …The Republic’s construction is that you can rebase midyear, refuse to apply the adjustment provision and thus avoid paying. So to say that that construction gives investors certainty is arrant nonsense …”

I am persuaded that there certainly could be non-fanciful arguments regarding commercial purpose. Further, though this is a complementary rather than a significant point in this context, the Claimants do pray in aid some fairly extensive factual matrix evidence. It was suggested that they are not entitled to rely on it in this context, because it is pleaded as relevant to the main issue, that of the approach to the Adjustment Provision. Although it does not matter for present purposes and I have certainly not relied on it thus far, I would not accept that submission; factual matrix evidence is relevant to the construction of the contract as a whole. Further, as I will explain below, although that factual matrix evidence may not actually be of direct relevance to these specific words, I do consider that the wider context of the case is not irrelevant.

Commercial purpose cannot therefore assist the Republic to a conclusion that its construction is right, for the purposes of the summary judgment application.

I would therefore conclude that the summary judgment application must fail, simply as a matter of the arguments on construction. But this conclusion is yet further underscored by the arguments on the wider issues in the case.

The core of the Claimants’ case is that the Republic should never have been in the position of considering the figures it did consider because the Securities required any change to the Year of Base Prices to be done in concert with a use of the Adjustment Provision. That, they say, applies to adjust Base Case GDP, and thus (because it is predicated on the figures which result) Base Case GDP Growth, for the purposes of the Conditions. If they are right about this, and about the construction of the Adjustment Provision, the Performance Condition would have been met. It is accepted for the purposes of this application that all of this is arguable.

It equally follows (and is accepted) that it is arguable that the Republic is not entitled to say that the Adjustment Provision can be disapplied by reason of its own failure/decision not to produce a Q4 figure in 1993 prices and not to continue publishing GDP figures based on 1993 base prices. The Claimants also have an equally arguable case that a new Year of Base Prices would not be validly adopted at all under the Securities if it was not made for proper purposes and in a way which is not irrational, arbitrary or capricious.

The Claimants say that given that background those claims, which sound in declaratory relief and specific performance, raise triable issues and the summary judgment application cannot, even in combination with the strikeout, effectively conclude the litigation.

The answer which comes from the Republic is that this is misconceived. It submits that it is apparent from the remedies which the Claimants seek that the essence of the Claimants’ case is that the Republic allegedly miscalculated the Payment Amount for Reference Year 2013. Thus, in addition to an order for payment, they seek declaratory relief that, inter alia, the Republic did not validly calculate the Payment Amount. That complaint necessarily involves substituting the Claimants’ own calculation for the calculation performed by the Ministry of Economy, which is precisely what the Binding Effect provisions preclude (absent bad faith, etc).

Thus it is said that the other remedies cannot have any content and that the Court will not make a declaration if no payment can be due even if there might conceptually be a residual discretion to do so.

While each side deployed this argument by way of an attempted knock-out blow against the other, it might equally be seen as an illustration of why, even if I had not reached the conclusion which I have, it might well be imprudent to decide the matter at this stage. The question of whether one is looking at a logically distinct exercise,

Further the question on operation of the Adjustment Provision is highly enmeshed with the argument about whether there was an implied term requiring the production of GDP statistics based on 1993 Base Prices. Is a conclusion that there is one exercise of calculation safely reached when one does not know if there is such an implied term? I consider that there are real dangers in splitting these points – and that was implicitly accepted by both parties. So as the Claimants pointed out, there is at least an oddity in concluding that there is one exercise of calculation if that exercise is qualified by a decision as to production of statistics which has no calculation element. And the Republic itself, while disclaiming any relevance of the Adjustment Provision to the question of construction, prayed in aid in the context of commercial purpose its approach to certainty derived from the underpinnings of the Adjustment Provision in the “baked in” percentages.

Therefore even had I not been persuaded that the Claimants have the better of the argument on the merits of the simple construction exercise, I would in any event have been quite clear that this was not a case where I should, as Mr Valentin urged me to do, grasp the nettle and grant judgment in his favour. This is not a case where leaving the point over would be a triumph of Micawberism.

In the circumstances the application for summary judgment fails. It was accepted that the strikeout, which concerns the feasibility of a defence only relevant if the Binding Effect Provisions apply, was contingent on success in relation to the summary judgment application, so it does not arise. However, I deal with it below briefly for completeness.

The essence of the Republic’s case was that the Claimants have not properly pleaded the necessary parts of the relevant defences.

It is easiest to start with manifest error as offering the lowest hurdle. Here the Republic’s point was simple. It submitted that “manifest error” in the Binding Effect provisions in this case requires the Claimants to plead and (if the matter were to proceed to trial) prove that an identified official of the Ministry of Economy made a calculation of the Payment Amount (or some other related calculation required under the Securities) on behalf of the Ministry of Economy, which was in error, and that error was obvious and easily demonstrable without extensive investigation.

The Claimants’ pleaded case alleges only that the Republic’s conduct amounts to manifest error because “its approach is manifestly inconsistent with the requirements of the Securities, including a refusal to apply the Adjustment Provision at all”. It is said that this is insufficient. There were effectively two limbs to this argument. The first was the formal requirement of pleading a natural person. On this I am not persuaded that the failure to plead a natural person renders the plea defective – and this element was rightly not really pursued. The second aspect of particularisation complained of (failure to particularise how the error was obvious) elides into the second point, that it was simply not possible for the hurdle to be met.

The second point really came back to the Republic’s underlying case on the Adjustment Provision, in combination with the submission that it could not seriously be suggested that an error that depends on the implication of an onerous new term into the Securities (that INDEC was required to continue to measure and publish or the Ministry of Economy was required to procure that INDEC continued to measure and publish GDP in outdated 1993 prices throughout the remaining life of the Securities) is one that was either obvious or easily demonstrable to any official at the Ministry of Economy, still less without extensive investigation.

The Republic says that:

However this argument seems to me to be impossible at this summary stage. If the Republic is wrong about the Adjustment Provision (which is plainly arguable), a question will arise as to the implication of the term – which it is accepted even at this stage is arguable. If that term falls to be implied it will be implied on the basis of necessity to give business efficacy to the contract. It may well not always follow that implication on this basis means a failure would be a manifest error. Doubtless the Republic would rely on the fact that the April 2014 Bank of America/Merrill Lynch thinkpiece to which I was directed indicates that the correct approach to rebasing for the Securities was a matter of some debate in the market. But at the same time, having spent some time considering the workings of the Adjustment Provision as it is seen by both parties, I can well see that it is possible that the combination of the test, and the nature of the issue in question would mean that a failure to comply with such a term would be held to result in a manifest error.

In particular I have been much assisted by the note which was produced overnight by the Republic’s team, which sets out with admirable clarity how it is said that the Adjustment Provision does not apply, and how it is said that if it does apply, the result is exactly the same as if it did not apply at all. Without burdening this judgment with detailed submissions and calculations which are not actually relevant to the live issues, what is apparent is that these are essentially points either of pure construction, or mathematical approach. These are types of issues which may well be amenable to a “manifest error” analysis if the implied term bites, and depending on the approach taken to the Republic’s secondary case.

I am quite satisfied that the argument that it would do so is well arguable. That view is only reinforced by the concession in the Republic’s skeleton that an obvious mistake would amount to a manifest error. That conclusion is not affected by the fact that on this hypothesis I would have concluded against the Claimants that the decision as to the Performance Condition (whether it is characterised as a decision not to perform the calculation of the Payment Amount or a decision not to apply the Adjustment Provision) was all a part of the calculation of the Payment Amount. The Republic’s approach to non-calculation of the Payment Amount hinges on its entitlement to

There was next to nothing between the parties on the law as regards the correct approach to strike out, or as to the requirements for pleading, and I shall not rehearse the common ground.

The issues between them were:

On the first point, this issue to some extent appeared to disperse as submissions progressed. To the extent it did not, I accept the Claimants’ submission that what is needed at the pleading stage is not a pleaded set of facts which lacks any other possible explanation than fraud, but rather the pleading of facts which, if proved, tilt the balance to fraud. This can be seen from the following authorities.

The first is the judgment of Lord Millett in Three Rivers DC v Bank of England (No.3) [2001] 2 AC 1, at 291E-H (HL):

“185.

It is important to appreciate that there are two principles in play. The first is a matter of pleading. The function of pleadings is to give the party opposite sufficient notice of the case which is being made against him. If the pleader means “dishonestly” or “fraudulently”, it may not be enough to say “wilfully” or “recklessly”. Such language is equivocal. ...

186.

The second principle, which is quite distinct, is that an allegation of fraud or dishonesty must be sufficiently particularised, and that particulars of facts which are consistent with honesty are not sufficient. This is only partly a matter of pleading. It is also a matter of substance. As I have said, the defendant is entitled to know the case he has to meet. But since dishonesty is usually a matter of inference from primary facts, this involves knowing not only that he is alleged to have acted dishonestly, but also the primary facts which will be relied upon at trial to justify the inference. At trial the court will not normally allow proof of primary facts which have not been pleaded, and will not do so in a case of fraud. It is not open to the court to infer dishonesty from facts which have not been pleaded, or from facts which have been pleaded but are consistent with honesty. There must be some fact which tilts the balance and justifies an inference of dishonesty, and this fact must be both pleaded and proved.”

The second is the judgment of Flaux J in JSC Bank Moscow v Kekhman [2015] EWHC (Comm) 3173 at [20]:

“The claimant does not have to plead primary facts which are only consistent with dishonesty. The correct test is whether or not, on the basis of the primary facts pleaded, an inference of dishonesty is more likely than one of innocence or negligence. As Lord Millett put it, there must be some fact “which tilts the balance and justifies an inference of dishonesty”. At the interlocutory stage, when the court is considering whether the plea of fraud is a proper one or whether to strike it out, the court is not concerned with whether the evidence at trial will or will not establish fraud but only with whether facts are pleaded which would justify the plea of fraud. If the plea is justified, then the case must go forward to trial and assessment of whether the evidence justifies the inference is a matter for the trial judge.”

Reference was also made to the judgment of Mann J in Gulati v MGN [2013] EWHC 3392 (Ch) where he declined to follow Moore-Bick LJ in ICI Chemicals TTE Training [2007] EWCA Civ 725 because Moore-Bick LJ had been dealing with arguments relating to construction and because:

“He was also not dealing with the familiar case in which a claimant makes an ostensibly sustainable allegation but acknowledges that the process of disclosure is necessary to make the case stronger or to have it investigated properly. It is a familiar state of affairs that a claimant is ultimately reliant on disclosure from the other side in order to bring his case home, particularly in cases where the nature of the wrong is such that the defendant's activities were covert so that, if the case is good, the defendant is likely to have a substantial amount of material in its hands with no equivalent in the hands of the claimant. Unless the prospects of getting disclosure are “fanciful”, the claimant is generally entitled to maintain its case in those circumstances. That is not to say that claimants are entitled to embark on speculative cases in the hope that disclosure will throw up something useful. The claimant must have more than that to start with, but the inability to make a full case without disclosure is not, in my view, a bar to starting the litigation in the first place…

Provided that there is enough to prevent them falling into the category of the purely speculative, the nature of the wrong alleged is such that the claimants will or may have little knowledge and evidence of their own at this stage and will need the benefits of pre-trial procedures in order to add to their case. There is nothing wrong with this. It is what disclosure (among other steps) is for. The alleged activities in this case were covert and, of their very nature, would be activities of which the victims would know little or nothing. Better evidence of what happened would lie with the defendant. There is nothing wrong with pleading a starting point, on an appropriate basis, and then expecting the case to become clearer after pleading and disclosure (if not the extraction of further information pursuant to a request).”

I consider that these authorities indicate that at the pleading stage what is needed is a pleaded case which, while it may contain allegations consistent with innocence, amounts to a case which if proved at trial, and making allowance where appropriate for the imbalance of information at the pleading stage, is capable of justifying a plea of fraud at trial. There need not be one killer fact or allegation; it is enough if those facts together are capable of tilting the balance.

The second issue between the parties concerned the meaning of bad faith. A contractual provision governed by English law that requires one party to the contract to establish that the other party has acted in “bad faith” is capable of being interpreted as requiring proof that the party acted with either:

This point is plainly in issue between the parties, but given the conclusion I have reached on summary judgment and manifest error I need not decide it now.

I would also note that as the Republic submitted, in practice there is unlikely to be any material difference between the application of these two definitions to the facts. The important point is that the Republic submits that even if, as the Claimants contend, “bad faith” is to be interpreted as requiring proof of an improper purpose short of dishonesty, this will require the Claimants to prove that the relevant decision was taken, or calculation made, by an official of the Ministry of Economy for a purpose which the official knew to be improper or incorrect, or suspected was improper or incorrect, but he or she carried on regardless of the consequences.

It follows that in the Republic’s submission, “bad faith” in the Binding Effect provisions in this case requires the Claimants to plead and (if the matter were to proceed to trial) prove that an identified official of the Ministry of Economy made a calculation of the Payment Amount (or some other related calculation required under the Securities) on behalf of the Ministry of Economy, which he/she knew to be improper or incorrect, or suspected was improper or incorrect, but carried on regardless of the consequences.

This was then used as the basis for a similar pleading point to that raised in relation to manifest error, with which I shall deal together with the same point, arising in the context of wilful misconduct.

There was little between the parties on the subject of wilful misconduct (which encompasses, and has the same essential elements as its close cousin, wilful neglect or default). Wilful misconduct by a party to a contract requires proof that the party either knew that it was acting in breach of duty or the law, or suspected that it was acting in breach of his duty or the law, but carried on without regard to the consequences and that person appreciated that his conduct created or might create a risk or harm in respect of the subject matter of the contract. Harm here is not in issue. The point on which the Republic placed emphasis was the need for conscious wrongdoing on the part of the relevant official.

Again there were two parts to the argument, the formal and the substantive. So far as the former is concerned, the Republic submitted that “wilful misconduct” or “bad faith” require the Claimants to plead that an identified official of the Ministry of Economy made a calculation of the Payment Amount (or some other related calculation required under the Securities) on behalf of the Ministry of Economy, which he/she knew to be a wrongful breach of his duty or a breach of the law, or suspected as much, but carried on regardless of the consequences. This level of particularisation is, the Republic says, lacking.

The second point is whether the pleaded case based on inference is sufficient to clear the hurdle delineated in the authorities.

As with the pleading point on manifest error, I am not unduly troubled by the first issue. It is entirely unrealistic to expect such a level of particularity, and I do not conclude that the authorities say that this is necessary. It may be necessary to prove conscious wrongdoing on the part of the relevant official, but it cannot be the case that the official needs to be identified in the pleading at the stage where the knowledge of the decision-making process is outwith the Claimants knowledge. There might be an argument that if advance disclosure had been given, such details should be pleaded; the position would then be more analogous to the ICI Chemicals case, but that does not arise here. The Republic has not acceded to requests for disclosure of its documents showing when, on what grounds and by whom the key decision was made, which might elucidate this subject. That is no criticism of the Republic, but it does at least potentially affect the degree to which the Claimants can be criticised for lack of particularity in their pleading.

The question then becomes one of whether the pleaded case on inference is sufficient. The Claimants’ pleaded case in respect of bad faith and wilful misconduct is founded on: (i) the twin assertions made in paragraph 45(a) and (b) of the Particulars, and (ii) the nine factual matters set out at paragraph 46(a) to (i) of the Particulars, which it is said lead to the inference that “the Republic did not have any or any real belief that the purported rebasing was being carried out on a sound statistical basis”.

The pleaded conduct which it is alleged constituted bad faith or wilful misconduct on the part of the Republic (Particulars, paragraph 45) is:

“(i)

deciding to switch to 2004 as Year of Base Prices and to immediately cease to publish or cause to be published data in 1993 prices, and/or (ii) in refusing to apply the Adjustment Provision, and/or (iii) the matters set out in paragraph 39 above”.

The question is whether this would suffice as a pleading of bad faith or wilful misconduct in the light of the authorities above. In the end I am persuaded that it would do so.

I do not accept that, as was submitted for the Republic, this allegation “necessarily involves an allegation of a largescale, interagency governmental conspiracy … to defraud the Claimants and all holders of the Securities”. It is quite possible for the decision to have been made by a key senior person, and then followed by more junior staff.

The Claimants acknowledge that their case on bad faith and wilful misconduct is necessarily an inferential one and it must also be accepted that a number of the particulars relied on are not particularly impressive. I do not propose to go through each subparagraph, as Mr Valentin did in submissions, but I can quite see that the arguments that there was nothing sinister in the use of 2004 as a new base year, or in not waiting for an updated census, have real force. So too do the arguments regarding what the Director of INDEC may or may not have said.

However, there is in the pleading a core of issues which, depending on the way the evidence comes out at trial, may be enough to justify such a conclusion. Part of this will involve the argument about the implied term. If that argument were to succeed, and the Court were to conclude that the term was so obvious and clear as to require to be implied to give business efficacy to the contract, that would be a point which (even absent manifest error) would provide food for thought as to how in those circumstances the Republic could have done otherwise. There are then the other issues to which I will allude below. Though none of these alone would have sufficient weight to justify a case, it cannot be said that, taken together, and depending on the strength of the case on each one, they could not tip the balance so as to justify an inference.

There will obviously be a vibrant issue as to the relevance of the IMF correspondence to the precise timing of the change. The Republic of course points to the chronology as indicating that the IMF had required remedial measures by September 2013. However, as Ms Prevezer noted in submissions, there is material for exploration as to whether the IMF were pushing for a change as soon as possible, which the Republic then actioned, or as to whether the IMF did not anticipate a change part way through the year. There may also be scope for argument about whether the IMF anticipated or the Republic had intimated that it would make a clean break.

There will also be an issue as to just how relevant the Republic’s past behaviour is, both in terms of quasi-admissibility and proper weight. However while I can see that there may well be grounds for saying that the judge should conclude that there is insufficient similarity, given the different nature of the alleged previous conduct, which related to understating CPI figures (as opposed to overstating GDP figures), to make it proper to say that the question of propensity is even open to be taken into account, it cannot be said that there is no scope for the argument. To use a criminal analogy, the past conduct

Adding to that the co-incidence of timing, the Republic’s financial difficulties at the time and the evidence that the payment of these bonds was attracting some controversy prior to the decision to rebase, I conclude that these facts do amount to a sufficient pleaded case to run the issues which are relied on. As I have noted I do not see some of the facts pleaded as adding much if anything to the argument, however given that the Republic explicitly makes the point that unpleaded facts cannot be relied upon to support an inferential case of dishonesty, an approach of “better safe than sorry” is plainly prudent.

Accordingly, I would also dismiss the strike out application.