The LDI (Liability-Driven Investment) Debacle, Derivatives and Systemic Risk There You Go Again!
The LDI (Liability-Driven Investment) Debacle, Derivatives and Systemic Risk There You Go Again!
The LDI (Liability-Driven Investment) Debacle, Derivatives and Systemic Risk There You Go Again!
The LDI (liability-driven investment) debacle, derivatives and systemic risk: there you
go again!
Vincenzo Bavoso1
Abstract
This article examines the role of derivatives in the context of the recent crisis in the UK pension
system, specifically in liability-driven investment schemes. It unveils derivatives’ role as
instruments that multiply and propagate losses among market participants once specific events
(worst case scenarios) materialise. This article questions the role of derivatives as risk-
management and risk-diversification tools, given their employment, in this case as in others
before, for speculative purposes. Critically, with speculation came higher than desirable levels
of leverage that these financial products elicited, causing in the process a series of systemic
concerns. While much criticism has been directed at post-2008 regulation, and particularly at
clearing and margin requirements, this article proposes a broader view of the problems posed
by derivatives in the different contexts of their applications. More specifically, attention is
drawn to legal doctrines that could be redeployed for the purpose of mitigating the speculative
nature of derivatives.
1 – Introduction
This article examines the liability-driven investment (LDI) crisis exploded in the UK
in the autumn of 2022, and specifically the role of derivatives in this crisis, as well as the
destabilising effects they more generally have on financial markets and the economy. The
volatility of the pension system at the end of 2022 offered an opportunity to re-evaluate risks
in the financial system, and particularly: a) how they are generated; b) the channels that
facilitate their propagation; and c) the segments of the financial system where they lie, hidden
from the scrutiny of market participants and more importantly regulators. Re-evaluating these
points is important, because it allows a much needed reflection on some of the functions of the
financial system, chiefly risk diversification and allocation, and some of the instruments that
1
Senior Lecturer in Commercial Law, Law School, University of Manchester. I presented an earlier draft of this
paper at a UCL workshop titled Defined Benefit Pension Schemes’ Liability Driven Investing Strategies, on 6
December 2022. I am thankful to the organisers and participants in that workshop for the very informative
discussions and feedback. I drafted this paper during a period of research leave as a visiting scholar at NUS Faculty
of Law. I thank the Centre for Banking & Finance Law at NUS for the kind hospitality and the great working
environment that was generously offered during my stay. Last but not least, I thank the reviewers for the very
constructive comments on a previous draft of this article. Errors are my own.
are traditionally designed to fulfil these functions: derivatives. Ultimately, this line of enquiry
aims to revisit the laws and regulations that are designed to stabilise these areas of financial
markets, with a view to appraise their appropriateness.
The instability at the heart of the pension system, and more specifically of LDI schemes,
as was experienced in the Autumn of 20222, is not a new phenomenon, but rather the
manifestation of an old problem that on this last occasion surfaced in a slightly different way
and context. It started, as it often does, with a political crisis and the announcement by the new
UK government of a mini-budget.3 This led to an unprecedented – and allegedly unpredictable
– decline in the value of gilts (UK sovereign bonds), which in turn had knock-on effects on
other segments of the financial system, such as the pension system.4
From a broader angle of analysis, this article looks at this episode of financial instability
as an instance of systemic concerns arising from the non-bank sector. It needs to be
remembered that before 2008, risks in the financial system were more clearly associated with
the banking system, or with channels of intermediation that were more directly connected with
large banks (as was the case for instance with the shadow banking system which was directly
connected with large banks through securitisation conduits and repo funding channels).5 After
2008, partly as a result of the greater regulatory scrutiny that has been placed on large financial
institutions, the locus of risk generation has migrated to a number of peripheral segments of
financial markets, which are less overtly connected with banks.6 This is precisely what leads
to the claims of “hidden” risks in capital markets, which have been voiced with more vigour as
a result of the LDI crisis.7
The main focus of this contribution is on the mechanics of the LDI debacle. In
particular, the thrust of this analysis is on the role that derivatives played in the structure of
these schemes. It is observed that the employment of a number of derivative products, such as
interest rate swaps and total return swaps, contributed to the initial success of LDI schemes,
making them a mainstream investment strategy in the pension system.8 The widespread use of
2
See Harriet Agnew, Adrienne Klasa, Josephine Cumbo, Chris Flood and Anjli Raval “How bond market mayhem
set off a pension ‘time bomb’”, Financial Times, October 8, 2022.
3
Ibid.
4
Ibid.
5
See BIS “Newsletter on bank exposures to non-bank financial intermediaries”, 24 November 2022.
6
See on this: Kaleeyna Makortoff “Bank of England to stress test hedge funds and private equity lending”,
Guardian, 13 December 2022; Brooke Masters “We haven’t reduced financial risk, just transformed it”, Financial
Times, 10 December 2022; Eric Platt, Kate Duguid, Tommy Stubbington, Jonathan Wheatley, and Leo Lewis
“Financial stability: the hunt for the next market fracture”, Financial Times, 5 December 2022.
7
Supra Agnew, Klasa, Cumbo, Flood and Raval, FT 2022.
8
See Dan Mikulskis “A Brief History of LDI”, Financial Times, October 5 2022.
derivatives led – as inevitably is the case in the context of present-day derivatives markets – to
an increase in the level of leverage in LDI schemes. It is precisely leverage that made for a
number of years these schemes successful (even during the global financial crisis period) and
profitable, attracting in the process layers of intermediaries and advisors.9 What is also
characteristic of derivatives though, is their capacity to multiply losses (as well as gains), once
specific events adversely affect the underlying reference assets. Events such as the mini-budget
in the context of this analysis, whose impact and likeliness had not been predicted or
incorporated into the mathematical models that underscore the use of these derivatives. It is
precisely the failure to incorporate such adverse risks that led to the instability of LDI schemes,
an instability that was aggravated by the level of leverage they were trading with.
This article ultimately contributes to existing scholarship that explores the function of
derivatives in financial markets, and their capacity to serve risk management purposes, as well
as speculative ones. While the recent episode in the UK pension system highlights wider policy
questions, this article investigates the more specific legal and regulatory facets the emerge from
it, reflecting in particular on how law can play an important role in constraining speculation.
The remaining of this article is developed as follows: Section two explains the rationale as
well as the processes (often referred to as financialisation) that led to the progressive
employment of derivatives in various areas of the economy, such as the pension system. This
part of the discussion also lays emphasis on the legal and regulatory changes that, over the past
forty years, elicited the wider application of speculative derivatives. Section three moves to a
closer examination of the mechanics of the specific derivatives contracts employed in LDIs,
namely interest rate swaps and total return swaps, uncovering in the process the capacity of
these contracts to create leveraged speculative bets. This leads to analysing in section four some
of the regulatory measures engineered after 2008 to mitigate the systemic risks that emanate
from derivatives. Emphasis here is laid on the mechanics of clearings and margin requirements,
which had the unintended effect of producing instability, more specifically by reinforcing
dynamics of liquidity crises, such as the one experienced in October 2022. Section five engages
with critical questions on the legal and regulatory treatment of derivatives. In their
contemporary shape, they have become instruments that propagate losses and instability in the
financial system, much against what conventional wisdom has traditionally postulated.
Building on the critique developed earlier in this article, this section discusses possible ways
9
As is explained at the end of section two, trustees operate in the interest of pension schemes and receive advice
from pension consultants. LDI schemes in turn are managed by asset managers who sell their product to the
pension scheme, liaising chiefly with pension consultants.
in which law and regulation could curb speculation and excessive risk-taking flowing from
derivatives. Section six concludes by putting forward a number of recommendations
10
More specifically, changes in the accounting standards (FRS17 and IAS19) in the 1990s had the effect of putting
pension deficits on corporate balance sheets, driving as a consequence corporate management to increasing
concerns over the size of a pension deficit. See Dan Mikulskis “A Brief History of LDI”, Financial Times, October
5 2022. For a critical and conceptual overview of these shifts, see Yuri Biondi and Marion Boisseau-Sierra
“Accounting for Pension Flows and Funds: A Case Study for Accounting, Economics and Public Finances”,
EGPA XII Permanent Study Group Public Sector Financial Management Workshop, Zurich-Winterthur
(Switzerland), May 7-8, 2015, Available at SSRN: https://ssrn.com/abstract=2606547.
11
John Plender “Lessons from the gilt crisis”, Financial Times, December 21, 2022.
12
Very briefly, in a pension scheme, assets are promises to pay made to the pension funds, and they comprise
investments such as shares, bonds, investments in real estate and so on; liabilities are promises to pay made by
the pension fund, and they chiefly are payments to the pension members. The shift discussed here is also the
reason why pension fund trustees started focusing primarily on hedging against inflation and interest rates, rather
than on maximising returns. Supra Plender 2022.
13
See BNY Mellon Investment Management “An Introduction to Liability Driven Investment”, November 2021,
p.5.
sovereign bonds have been key hedging instruments in this respect.14 More specifically, with
inflation-linked bonds for instance15, interest rates fluctuations affect the value of bonds in the
same way they affect pension liabilities.16 Investing in bonds on the assets side of pension
schemes soon became insufficient though, for two main reasons. First, bonds that fulfil the
necessary hedging requirements are finite in nature (typically, sovereign bonds have those
characteristics), and second, holding bonds on balance sheet can be expensive for pension funds
and inefficient from a capital point of view. In essence, buying bonds entails for a pension
scheme sacrificing investments into growth assets (such as equity most typically), which is a
strategy that most schemes cannot afford.17
Together with the scarcity of bonds, capital efficiency soon became a priority for
pension funds – for a number of reasons that are further explored later in this paper – and this
is where the use of derivatives in LDI schemes came to play a central role in the pension
industry. The idea here was for pension funds to adapt their portfolio of assets and gain a
synthetic exposure to bonds through derivative products18, whereby the capital efficiency of
derivatives would enable the pension fund to invest at the same time in other assets for the
purpose of generating growth and improving funding levels.19 This is precisely what LDI
schemes became prominent for, namely managing liability risks and generating investment
returns for members. In other words, the fundamental aim of the scheme is to create a hedge
against liability risks, and have assets and liabilities that rise or fall in value together, so that
the funding level of the pension fund remains stable over time.
The capital efficiency that is at the heart of LDI schemes entails investing in a range of
partially funded instruments20 where the exposure to a specific asset (or an asset class) is
achieved synthetically, and with limited capital commitment.21 This is an investment pattern
14
Essentially the price of bonds behaves as a pension scheme’s liabilities, therefore by holding a portfolio of
bonds on the assets side of its balance sheet, a pension fund can gain protection against changing interest rates
and yields. In other words, variations in bond yields offset the variation in the pension fund’s liabilities. See BMO
Global Asset Management “Liability Driven Investment Explained”, 2018, p.15.
15
That is, bonds that provide a series of payments linked to inflation as applied to both the coupon and the notional
of the bond. Supra BMO Global Asset Management, 2018, p.16.
16
Ibid, p.17. A fall in interest rates will increase the value of a pension scheme’s liabilities, and the bond will also
increase in value. In a similar way, an increase in inflation will see a corresponding increase in the value of
liabilities as well as the value of an inflation linked bond.
17
The argument here is that holding bonds on balance sheet may prevent the pension fund from investing in more
growth oriented assets. Supra BMO Global Asset Management, 2018, p.17; BNY Mellon Investment
Management, 2021, p.8.
18
This synthetic exposure was achieved by entering into a number of contracts, typically interest rate swaps, total
return swaps, and gilt repos.
19
See BNY Mellon Investment Management, 2021, p.5-7.
20
See fn. 18.
21
Supra BNY Mellon Investment Management, 2021, p.5-7.
that is made possible through the use of derivatives, as will be explained in the next section,
because essentially the underlying assets that the pension fund is seeking exposure to are never
acquired and never change hands. It follows that, instead of purchasing the full price of say
gilts, the pension fund will only have to pay a fee to enter into a derivatives contract that
achieves exposure to that amount of gilts.
The rationale for employing derivatives in the context of LDIs seemed compelling, both
with respect to the specific context of pension funds, and in the wider realm of ever more
developed capital markets. It is worth reminding here that the mass application of credit
derivatives in LDIs coincided with a period of regulatory changes that peaked in the early
2000s22, after which the market for credit derivatives grew exponentially, initially in the US
and the UK, and then globally.23 Regulatory changes in that period were essentially aimed at
liberalising these markets, and this was done primarily by removing restrictions on certain
types of derivatives that had been historically considered speculative. The remaining of this
section draws a close analysis of the legal and regulatory changes that propelled the expansion
of speculative derivatives.
22
It needs to be noted that the development of derivatives, from relatively niche products based on commodities
markets, to products based on financial assets and index, started taking place in the second half of the 1970s,
following the collapse of the Bretton Woods agreements (and the fixed exchange rate that was a key part of it),
and the period of financial liberalisation that ensued. On this, see Satyajit Das “Traders, Guns and Money”, FT
Publishing 2020, p.40; see also Donald MacKenzie “An Engine, Not a Camera”, MIT Press 2008.
23
See on this, Adam Chambers “Credit derivatives: BBA and Isda report explosive derivative growth”,
Euromoney, 29 September 2006, showing the massive increase in the value of global derivatives markets between
1996 and 2006. See also on this Bank for International Settlements “Press Release: The global OTC derivatives
market at end-June 2001”, BIS, 20 December 2001.
24
See Henry Dewey “A Treatise on Contracts for Future Delivery and Commercial Wagers, Including “Options”,
“Futures” and “Short Sales”, Baker, Voorhis & Co Publishers, 1886, Preface.
to validate this claim by way of analogy, on the ground that insurance too plays a vital economic
function. In an insurance, Stout observes, the wager results in a mutually beneficial transaction
for both parties who are able to hedge risks.25 If insurance can be seen as a commercially
beneficial form of betting, merely speculative betting is something that does not lead to a
mutually beneficial exchange, because it creates instead illusions among risk averse investors,
who become exposed to new risks without any corresponding increase in the potential return.26
A more straightforward take on this dichotomy is that provided by Hudson, who proposes a
distinction between hedging contracts, which revolve around the management of risks, and
speculative contracts that essentially involve betting on the performance of underlying assets.27
Given these rather differing characterisations, it is important to explore the common
law approaches to this important question. Courts, both in the US and the UK, developed
historically principles and criteria that had the effect of constraining certain applications of
derivatives that were considered aligned with a gambling rationale.28 An important principle in
this sense came from the US with the old common law rule established by Irwin v. Williar, a
case dealing with the sale of future goods.29 The court identified the delivery, or at least the
deliverability of the underlying assets, as the discerning element between the two contractual
functions, namely speculation and hedging.30 The delivery in essence represented a proprietary
interest of the contracting parties in the underlying physical assets (that being a quantity of
wheat for instance, or in the case of LDI schemes, a number of gilts). As a result of this
precedent, contracts where the parties were not expected to take delivery of the underlying
referenced assets were considered by courts to be mere wagers – speculations on the value of
those assets in the future – and as such were not enforced by courts.31 Importantly, an exception
25
In the example of a fire insurance policy, the wager is related to the occurrence of fire. The aim of taking
insurance in this case is for the house owner to offset the loss that occurs in the event of fire. See Lynn Stout
“Derivatives and the Legal Origins of the 2008 Credit Crisis”, Harvard Business Law Review, Vol.1, 2011, p.7.
26
Ibid, p.8. Stout uses synthetic CDOs as an example of a speculative bet, where the creation of new risks leads
to rent seeking.
27
See Alastair Hudson “The Law of Finance”, Sweet and Maxwell, 2013, p.1176-1179.
28
Supra Stout 2011, p.8, who explains how these criteria were rooted in a social and welfare rationale. This
becomes more evident later in this article when the analysis looks at the relevant provision of the UK Gaming Act
1845.
29
110 U.S. 499, 508–09 (1884).
30
This is explained in detail by Lynn Stout “Derivatives and the Legal Origin of the 2008 Credit Crisis”, 1 Harvard
Business Law Review, 2011, p.11-12.
31
This approach was embedded in Irwin v. Williar 110 U.S. 499, 508–09 (1884). In this case the court reiterated
that a contract for the sale of goods to be delivered in the future would be valid even if the seller had no goods
(nor any means of getting the goods other than buying them on the market). This type of contract would only be
valid if the parties agreed for the goods to be delivered by the seller and paid by the buyer. Conversely, if the real
intent under this type of contract was to speculate on the rise or fall of the price of goods, and the goods were not
to be delivered (with one party paying the other the difference between the contract price and the market price of
the goods), then the transaction would be deemed a wager and as such would be void before courts.
to this this rule existed, which confirmed the overall attitude towards protecting hedging while
penalising speculation. Notwithstanding the absence of deliverability, courts would still
enforce the contract if one party had a pre-existing economic interest in the underlying goods
that would be damaged by the same event that would allow profiting under the contract. The
rationale for this exception is (again) very similar to the context of insurance law, where the
insurable interest represents a condition for enforceability, and which is relevant to this
discussion later in this article.32
Similar orientations emerged from early UK cases, which lay the foundations of the
modern understanding of contracts for difference (CfD), as derivatives are often referred to in
statutory instruments (this point is reprised in the next section). In Universal Stock Exchange
v. Strachan, the House of Lords held that contracts for difference, that is where participants
speculate on the performance of the underlying price, were void as they were considered
gaming contracts.33 An elaboration of this principle came in much later cases. In City Index v.
Leslie it was maintained that the purpose of the contract (in that case spread bets) was not
limited to a hedging transaction, but the contract for difference structure could serve a
speculative function.34 A more complete definition of contracts for difference came with the
decision in Larussa-Chigi v. Credit Suisse First Boston, where, in the context of foreign
exchange transactions, the deliverability principle (as was established with Irwin v. Williar)
was reiterated as the distinctive feature of CfDs.35
The adversity towards gambling, often used by courts to justify their approach towards
certain contracts, was informed by specific gaming legislations. In the UK, the social and moral
attitude towards gambling was embedded in a piece of Victorian legislation, the Gaming Act
1845. This provided general prohibitions against a wide range of activities that were
characterised as gambling or wagers, among which financial contracts, which were not
recognised by law.36 In the US, and specifically in New York, a similar legislative source
provided that all wagers and bets made to depend on an unknown or contingent bid shall be
unlawful.37 Notwithstanding these old prohibitions, modern gaming legislations no longer
32
Supra Stout 2011, p.12.
33
[1896] AC 166, para18-22.
34
[1992] QB 98.
35
[1998] CLC 277. In particular, it was emphasised that in a CfD, the parties’ intention is only to settle the profit
between the movement of the two amounts, as opposed to taking physical delivery of the gross foreign amount.
36
Section 18 of the Gaming Act included a wide range of prohibited gambling activities, among which also
financial contracts that resulted in gambling. Such contracts, according to the Act, would be void and not
recognised by law.
37
See Section 5-401 of the General Obligation Law of the State of New York.
apply in the context of derivatives markets, and the related restrictions were progressively
relaxed38, only to be completely removed between the mid-1980s and the late 1990s.
In the UK, the Financial Services Act 1986 (section 63) had the effect of limiting the
scope of section 18 of the Gaming Act 1845, creating an exemption for investment activities,
such as derivatives, which could be entered into as pure wagers, for speculative purposes.39
This legislative approach was consolidated at common law, through a number of cases
illustrated below. In the US, courts have tended to dismiss arguments that derivatives should
be considered as gambling and thus void.40 In the UK, Morgan Grenfell & Co Ltd v. Welwyn
Hatfield DC provided a useful reference to the question of what amounts to a speculative
contract – in that context, an interest rate swap.41 The court stressed that in the context of
derivatives entered into by parties in capital markets, it would be fair to assume that such
transactions were not wagering, but commercial and financial contracts, and that as such they
should be enforceable. It added that, given the inherently speculative nature of an interest rate
swap, the assumption could be rebutted if the main purpose and interest of the contract was
indeed a wagering one. However, in the context of this case the wagering element was
considered subordinate, and as a result, the swap could not be considered a wager under section
18 of the Gaming Act.42
An interesting aspect that emerges from UK cases, particularly in light of decisions
such as Hazell v. Hammersmith and Fulham LBC43, is the widespread assumption that
derivatives, and swaps in particular, were indeed used for speculative intents by local
38
For a detailed and multifaceted account of this process, see Donald MacKenzie “An Engine, Not a Camera”,
MIT Press 2008; Donald MacKenzie “Material Markets: How Economic Agents are Constructed”, Oxford
University Press, 2009.
39
For an overview of the “Big Bang” in the City of London, see Laurence Gower “”Big Bang” and City
Regulation”, The Modern Law Review, 51, 1, 1988. Today, the Gambling Act 2005 provides for the licencing of
gambling businesses, but those regulated by the Financial Conduct Authority are exempted. See Philip Wood
“Regulation of International Finance”, Sweet and Maxwell 2019, p.234.
40
See in this sense, Career Life Insurance Co v. Morgan Guaranty Trust Co of New York US Dist Lexis (SDNY
2003).
41
[1995] 1 All E.R. 1.
42
Ibid. Also of relevance, as a tentative basic definition, is the one provided in Carlill v. Carbolic Smoke Ball Co
[1892] 2 QB 484, where the court held that “it is essential to a wagering contract that each party may under it
either win or lose, whether he will win or lose being dependent on the issue of the event”. See also as a valid
precedent on this issue Hazell v. Hammersmith and Fulham LBC [1992] 2 A.C. 1, [1991] 1 WLUK 792, where
the court illustrated both the commercial and the speculative rationale of an interest rate swap, and concluded that
the local authority's purpose was “more akin to gambling than insurance”.
43
[1992] 2 A.C. 1, [1991] 1 WLUK 792.
authorities (seeking to maximise their debt exposures while still generating income), and that
this was a valid purpose.44
To complete this discussion, more critical legislative shifts occurred in the US, where
the lobbying power of the fast changing derivatives industry45 led to reduced regulatory powers
of the CFTC (the Commodities Futures Trading Commission, the US main regulator in the
field of derivatives), and eventually to the passing of the Commodities Futures Modernization
Act 2000.46 This had the effect, among other things, of dispensing with old common law
distinctions between speculative and hedging contracts, and gave legal certainty to all
speculative derivatives traded over-the-counter (OTC) by a wide range of eligible
counterparties.47
2.2 – Dispensing with proprietary interests: cash settlement and insurable interest
The discussion conducted in the previous section shows that both in the UK48 and in
the US49 derivatives progressively developed as instruments that could facilitate speculation
on price variations of underlying assets. This trend was grounded on the widespread acceptance
and application of contractual structures where counterparties had no proprietary interest in the
underlying assets, and instead of acquiring them could simply enter into a virtual speculation.
This is the essence of the contractual structure that is known as cash settlement. It means that,
given that neither party is interested in taking delivery of the assets, the contract is settled with
the payment of the difference between the price of the assets at maturity (the market price) and
the price fixed in the contract (the strike price).50
In essence, cash settlement is the contractual structure that elicits speculation, as parties
effectively bet on the price fluctuation of specific assets in the future. This is achieved by
creating a right to pay or receive a cash return on the performance of a notional amount of
44
Ibid. The House of Lords concluded that if the swaps were within the powers of the local authority, there would
be no objection to their validity and enforceability (the main question here was whether local authorities had acted
ultra vires).
45
This process is very well explained and documented in Stout 2011, part IV; and MacKenzie 2008.
46
Pub. L. No. 106-554, 114 Stat. 2763 (2000).
47
Supra Stout 2011, p.21, where she explains in details how the 2000 Act not only made all derivatives legally
enforceable but it also freed the market of the oversight of the two main regulators, the Securities and Exchange
Commission and the Commodity Futures Trading Commission.
48
This transpires particularly from the decisions in the context of the swaps purchased by local authorities, as
illustrated in the previous section, and particularly with Hazell v. Hammersmith and Fulham LBC.
49
The shift in the US started in the mid-19th century, when traders started exchanging abstract contracts based on
the market prices of various assets – this shift was facilitated by contractual practices whereby trades were
perfected not by actual delivery but by set-off, achieving in other words the same economic function of contracts
for differences. See for an explanation Stout 2011, p.14.
50
This settlement structure is also at the heart of the contracts commonly known as contracts for difference (CfD).
10
money, whereby that notional amount mimics the performance of a given market.51 It is useful
to stress that the widespread acceptance of these contractual forms is embedded in the language
of statutory provisions, which refer to cash-settled financial derivatives or contracts for
differences.52
The applicability of derivatives for speculative purposes is premised on the
development of another important legal doctrine, which brings back the analogy with insurance
law: the insurable interest. Its basic tenet implies that the insured benefits from the preservation
of the subject matter of the policy (the asset that is insured), or suffers a disadvantage from its
loss.53 This is a common law doctrine that typically applies to insurance contracts, whereby an
“insured must derive a benefit from the continued existence of the insured entity. Absent an
insurable interest, an insurance contract may be void”.54 It could be argued that many
derivatives accomplish an insurance-like economic function (particularly swaps), and the
application of the insurable interest could contribute to further defining the difference between
hedging (insurance function) and gambling. The concept of insurable interest would in other
words substantiate the idea of a mutually beneficial transaction between parties that are seeking
to hedge risks – the disadvantage from the loss of the insured asset reinforces the idea of the
proprietary interest that the insured has over the asset.
Despite the insurance-like function that many derivatives have (think about the payment
structure under a swap), credit derivatives are not characterised as insurance contracts, and as
a result the absence of an insurable interest cannot be seen as impairing the validity of
derivatives. This authority is rooted in the opinion provided in 1997 by Robin Potts QC, upon
the request of the International Swaps and Derivatives Association (ISDA).55 Potts held that
credit derivatives were not insurance contract for the purpose of the existing legislation (back
then the Insurance Companies Act 1982, after that the Financial Services Market Act 2000)
and at common law, chiefly because they are designed to pay out upon the occurrence of a
51
In this sense, see Hudson 2013, 1178.
52
See for instance the Financial Services Act 2000, para.15-25.
53
See Gary Meggitt “Insurable Interest – The Doctrine that Would not Die”, Legal Studies, Vol.35 N.2, 2015,
p.281. Meggitt also explains that there are further questions with this doctrine, in relation with the time when in
needs to exist, and the different contexts and types of insurance policies.
54
See Alexander Charap “Minimizing Risks, Maximizing Flexibility: A New Approach to Credit Default Swap
Regulation”, 11 Business & Securities Law, 127, 2011, p.156. For an illustration, see also Oskari Juurikkala
“Credit Default Swaps and Insurance: Against the Potts Opinion”, Journal of International Banking Law and
Regulation, Volume 26, Issue 3, 2011.
55
Opinion prepared for the ISDA by Robin Potts QC, Erskine Chambers, 24 June 1997.
11
credit event, irrespective of whether the protection buyer suffers a loss 56, and also because of
the absence of insurable interest to be protected in the context of a swap.57
56
Ibid, para 5.
57
Ibid, para 5.
58
See Alan Greenspan “Risk Transfer and Financial Stability: Remarks by Chairman Alan Greenspan to the
Federal Reserve Bank of Chicago’s Forty-first Annual Conference on Bank Structure”, May 5 2005, available at
www.federalreserve.gov/Boarddocs/Speeches/2005/20050505/default.htm.
59
Ibid. Critically, Greenspan was a firm believer in the capacity of the financial system to manage issues of
counterparty credit risk, which may manifest themselves particularly in the context of the OTC market. In
particular, he assumed market participant capable to evaluate each other’s creditworthiness and withdraw from
transacting with parties they deemed too risky.
60
The same efficiency rationale underscores and explains transactional developments such as the move towards
cash settlement and the trading of derivatives OTC, explored earlier in this section.
61
Warren Buffet is CEO and majority shareholder of the investment firm Berkshire Hathaway Inc.
62
See Pablo Triana “Buffett’s derivatives bonanza doesn’t prove that Taleb is wrong”, Financial Times, 6 July
2023, where the above mentioned letter is referenced. At the end of that letter, Buffet uttered the famous lines
where he stated that “derivatives are financial weapons of mass destruction, carrying dangers that, while now
latent, are potentially lethal”.
12
Notwithstanding those earlier concerns, and while witnessing the sharp and unchecked
expansion of derivatives, Buffet also noted the rather helpless position of regulators, who were
unable at the time (as they seem to be today) to monitor the risks emanating from these
contracts.63
It was also widely believed that risks flowing almost axiomatically from the opacity of
the OTC market, primarily counterparty credit risk, could be mitigated through the device of
market discipline.64 Without delving into the tenets of market discipline, which would fall
beyond the scope of this paper, it suffices to say that in this context it resulted in the belief that
the financial system (and specifically entities within it) could manage issues of counterparty
credit risk without the need of external regulatory intervention. This orthodoxy was rooted in
the idea that market participants in financial markets would be capable to monitor and evaluate
each other’s creditworthiness, and whenever necessary withdraw from transacting with parties
deemed too risky.65
In the more specific context of the pension system, credit derivatives proved hugely
profitable as part of LDI schemes, because in many instances successful LDI strategies
achieved the goal of growing the value of assets faster than liabilities – something that was
probably beyond expectations and even unnecessary given that the original goal was simply to
have a sufficient level of funding to pay off liabilities. Moreover, LDIs performed well during
the turbulent years of the GFC, and between 2014 and 2018 they became the mainstream
investment pattern in pension funds.66 With the growth of this sector, it is inevitable that it
became permeated by a plethora of intermediaries and advisors, whose role came to the fore in
the context of the LDI debacle.
Briefly, LDI schemes are managed by asset managers, who essentially design,
recommend and manage the investment strategy within LDIs, in the interest of the pension
scheme (their client).67 Inside pension funds, trustees and their consultants determine the
63
Ibid.
64
For a discussion on market discipline, see Emilios Avgouleas and Jay Cullen “Market Discipline and EU
Corporate Governance Reform in the Banking Sector: Merits, Fallacies, and Cognitive Boundaries”, Journal of
Law and Society, Volume 41, Issue1, 2014. See also Vincenzo Bavoso “Basel III and the Regulation of market-
based Finance: The Tentative Reform”, New York University Journal of Law and Business, Vol.18, N.1, 2021.
65
Supra Greenspan 2005; see also Andrew Crockett “Market Discipline and Financial Stability”, BIS Speech, 23
May 2001, available at https://www.bis.org/speeches/sp010523.htm. Crockett warns that for it to be effective,
market discipline rests upon a number of conditions, namely that market participants have sufficient information
to reach informed judgements, that they have the ability to process information, that they have the right set of
incentives, and that they have the mechanisms to exercise discipline.
66
Supra Mikulskis, FT 2022.
67
Russell Investments “Pension risk management strategies”, available at
https://russellinvestments.com/us/solutions/defined-benefit/liability-driven-investing.
13
objectives and the level of exposure to LDI strategies.68 Key questions revolve around the
nature of the advice provided by pension consultants (as well as the service of asset managers),
the quantitative models underpinning it, and the rationale for the level of leverage that they
were advocating. Some of these questions are explained in the next section which focuses more
specifically on the mechanics of credit derivatives.
3 – The use of derivatives in LDIs: Interest rate swaps and total return swaps
The fundamental aim of LDI schemes is to create a mirror image on the pension
scheme’s assets of the scheme’s liabilities.69 Typically, interest rate swaps (IRS), total return
swaps (TRS) and gilt repos, are used to accomplish this function. As this paper is concerned
with the use of derivatives in pension schemes, gilt repos are left out of this analysis.70
In its essence, a swap is a bilateral contract whereby two counterparties agree to
exchange two pecuniary obligations. A definition of a swap contract was provided in Hazell v.
Hammersmith & Fulham LBC, where the court, dealing with an IRS, characterised it as an
“agreement, whereby each party agrees to pay the other at a specified date, an amount
calculated by reference to the interest which would have accrued over a period of time on the
same notional principal sum, assuming different rates of interest are payable in each case – one
rate being fixed (10%), and the other floating (equivalent to the six month LIBOR). If over that
period of the swap, LIBOR is higher than 10%, then the counterparty agreeing to receive
interest payments in accordance with LIBOR will be entitled to receive more than the
counterparty entitled to receive fixed interest payments”.71
In the context of pension schemes, the exchange of the two payment flows is between
the pension scheme as buyer of the swap, and an investment bank acting as seller of the swap.72
As introduced in the previous section, a pension scheme will typically enter into an IRS in
order to receive a fixed interest payment (of say 3.5%) for a specific time period (say 20 years),
68
BlackRock “LDI strategies: setting the record straight”, available at
https://www.blackrock.com/uk/solutions/insights/liability-driven-investing.
69
Supta Mikulskis, FT 2022.
70
See BNY Mellon “An Introduction to Liability Driven Investment”, November 2021, p.11. Bearing in mind
that while the contractual structure of repos differs from that of swaps, the economic function of the two
transactions converges and both are partially funded instruments.
71
1991 1 All ER.
72
Since the 1980s large financial institutions (dealer banks) started warehousing derivative positions, for the
purpose of matching long and short positions entered by counterparties, whereas previously they would simply
broker the two counterparties’ positions. Effectively, this led to the progressive standardisation of trades, centred
on dealer banks.
14
and over a notional asset, which in this case is a gilt (say £100). Over the same specific time
period, the pension scheme pays a variable interest on the same notional amount.73
Total return swaps (TRS) are bilateral contracts where two counterparties exchange the
returns (in essence the performance) from a specific financial asset. Under the contract, one
party (the investor) makes payments based on a set rate, while the other (a bank holding the
financial asset) makes payments based on the return of the underlying asset.74 In the context of
LDIs, TRSs achieve the exchange between a floating rate on the one hand, and the cashflow
from the gilt on the other hand. Again, for the pension scheme the cashflow from the TRS
reflects that of the underlying asset.75
These contractual structures are designed to allow pension schemes to gain exposure to
the payment stream flowing from gilts. With a derivative though, the underlying asset (the gilt)
does not change hands, because as explained previously, a synthetic exposure is instead
achieved.76 What is also important to highlight with respect to swaps is that when the contract
is entered into neither side of the swap is more valuable than the other. The value of the swap
(and the cost of the two pecuniary obligations therein) is chiefly affected by shifts in interest
rates. In the above example, if interest rates were to rise above 3.5%, the variable leg of the
swap would be more expensive than the fixed one, meaning that the pension fund would lose
money on the swap. Conversely, a decline in interest rates would make the swap profitable.77
To put this analysis further in the context of the LDI crisis, it needs to be remembered
that by 2019 the overall value of LDIs had surpassed £1tn, and at that stage the amount of gilts
in circulation was clearly not sufficient to provide hedge against all the pension schemes.78
That partly explains the increase in the level of leverage across the LDIs. While they were
always designed to cope with a certain degree of rising interest rates, the scale and speed of the
73
Supra BMO Global Asset Management, 2018, p.19. The variable interest is typically calculated by reference to
a 3 or 6 month lending rate, such as LIBOR, or what replaced it in the UK, SONIA (Sterling Overnight Index
Average).
74
See for an illustration Corporate Finance Institute “What is a Total Return Swap (TRS)?”, available at
https://corporatefinanceinstitute.com/resources/knowledge/finance/total-return-swap-trs/. For another example,
based more specifically on the case of hedge fund Archegos, the hedge fund enters into a TRS with a dealer bank
on a certain stock, as the underlying asset, for a notional amount ($10ml). The dealer bank purchases the notional
amount of the specific stock, while the hedge fund posts an initial margin (15%) that is a percentage of the notional
amount. If the price of the underlying stock increases by 10%, the hedge fund’s long position would be rewarded
and it would receive variation margins from the dealer bank ($1ml). In case of a decline in the value of the
underlying stock, the hedge fund would have to pay variation margins to the bank ($1ml). See ESMA “Leverage
and derivatives – the case of Archegos”, May 2022, p.4.
75
Supra BMO Global Asset Management, 2018, p.20.
76
This ties with the idea of capital efficiency referred to earlier. To gain exposure to a £100 gilt, the LDI does not
need to invest £100 for the purchase of the bond; instead it only pays a fee to enter into the swap, and the movement
in the value of the swap reflects that of the underlying bond.
77
Ibid, p.18.
78
Supra Mikulskis, Financial Times, 2022.
15
events in October 2022 put huge pressure on the pension system, and beyond. Once again, as
was the case in 2008, but also in more recent episodes such as the demise of US hedge fund
Archegos79, the employment of derivatives accelerated problems of leverage,
interconnectedness and systemic risk.
Something about leverage needs to be explained at this stage of the analysis. In simple
terms, leverage represents the percentage of debt that is used in an investment, relative to
equity.80 In financial markets leveraged investments have traditionally been welcome due to
their capacity to amplify returns on equity81, and to enable accessing investment opportunities
with limited amounts of capital (exactly what was said earlier in the context of LDIs). In this
sense, leverage can be seen as value enhancing and an element that contributes to creating deep
and complete financial markets.82 The relevance of leverage to this paper’s analysis hinges on
the widespread employment of derivatives aimed at enhancing leveraged investments. In other
words, while initially leverage was embedded in LDI strategies to maximise pension schemes’
funding levels, it became a way to simply amplify returns. This was possible due to the
contractual aspects of derivatives that were discussed earlier in this paper, because investors
can gain exposure to certain assets without actually owning them.83
With the unfolding of the events of October 2022, the drawbacks typically associated
with high levels of leverage became far too visible. The LDI meltdown was caused by a higher
than expected rise in the interest rate of gilts, which was in turn a reaction to the UK
Government announcement of the mini-budget and the subsequent fall in the value of gilts.84
It was explained earlier that under the swaps arranged in the context of LDIs, an increase in
interest rates would cause the derivative contract to be out of the money for the pension scheme.
The level of leverage that pension schemes were engaging with made the fluctuations in the
bond market, and their positions in it, far more problematic. More specifically, the declining
79
See on this ESMA “Leverage and derivatives – the case of Archegos”, May 2022.
80
A straightforward example is that of the purchase of a house with a mortgage. In a situation where the house is
valued £100, the buyer borrows £90 from the bank and pays a deposit of £10, the level of leverage can be
represented by a debt to equity ratio, in this case 9:1. See for a critical discussion on leverage Margaret Blair
“Financial Innovation, Leverage, Bubbles, and the Distribution of Income”, 225 Review of Banking & Financial
Law, Vol.30, 2010-11.
81
Following the example in the previous footnote, if the value of the house increases by 10% after 5 years, to
£110, the buyer will have realised a return on the initial equity investment of 100%. If the house had been bought
with no leverage, the return on equity would have been 10%.
82
On market completion as a policy goal, see for instance Andre Sapir, Nicolas Veron and Guntram Wolff
“Making a reality of Europe's Capital Markets Union”, Bruegel Policy Contribution No. 2018/07.
83
Satyajit Das explains how the structure of derivatives and their use to speculate on the value of underlying assets
could lead to huge gains, but also to the multiplication of losses. “Traders, Guns and Money”, 2020, p.37-38.
84
See John Ralfe “Investigation needed to hold those behind the pension crisis to account”, Financial Times, 13
October 2022.
16
value of gilts directly affected the value of leveraged positions, which also collapsed in value.
This in turn triggered the increase in the amount of collateral that pension schemes had to post
(due to the clearing obligation discussed in the next section).85 To meet these collateral calls,
pension schemes started selling their most liquid assets, the gilts, further exacerbating the
collapse in their value and the ensuing liquidity crisis.
As is typical of financial cycles, high levels of leverage were hailed for the huge gains
they were eliciting, but as the tide turned in October 2022, and “unforeseen events” caused a
dramatic rise in interest rates, leverage became “the cause” of the LDI collapse.86
The paradox observed in the context of the LDI crisis is that a scheme that was
conceived to manage risks became the very source of risk propagation. While this is a feature
that has become intrinsic of derivatives, due to the legal characteristics discussed earlier,
questions of risk propagation have taken a new turn in the post-2008 years.
85
Ibid.
86
The LDI collapse was of course attributed to a mix of different causes, such as the political instability, both in
the UK and internationally, the post-pandemic debt overhang, and monetary policies. As is discussed later in this
paper, asset managers advising pension schemes on LDI strategies claimed that what happened in October 2022
was an unpredictable scenario that could not be modelled. See Mikulskis, Financial Times, 2022.
87
John Plender “Lessons from the Gilt Crisis”, Financial Times, December 21 2022.
17
Both sets of problems were addressed by introducing the so-called clearing obligation
for OTC trades, which essentially mandated that derivatives contracts should be cleared by a
central clearing counterparty (CCP). A definition of clearing was provided in the context of
EMIR (European Market Infrastructure Regulation)88, as the process of establishing positions
(which include the calculation of net obligations) and ensuring that financial instruments are
available to backstop the exposures that arise from the same derivatives positions. In essence,
the clearing obligation changes the bilateral nature of derivatives with the interposition of the
CCP which stands between the two counterparties for the purpose of guaranteeing the
performance of the two respective obligations. This was conceived to mitigate counterparty
credit risk and reduce as a result the occurrence of a systemic crisis.89
Together with the clearing on CCPs, post-2008 regulation also introduced trade
repositories (TRs) which serve as central data collectors and maintain records on derivatives
positions.90 This is precisely the measure aimed at mitigating the second problem signposted
earlier, that is the lack of transparency in derivatives markets and the ensuing risk of systemic
stability.
It should be noted that in the US similar measures were implemented by the Commodity
Futures Trading Commission (CFTC) that issued final rules on clearing requirements under
section 723 of the Dodd-Frank Act.91 These require certain classes of derivatives to be cleared
by clearing organisations that are registered with the CFTC. In both regimes, clearing
obligations come with a number of exceptions.92
An important consequence of the clearing obligation is the attendant requirement that
counterparties have to post initial and subsequent collateral (what is also called margin).93 The
function of margin requirements is to absorb potential future losses, and thus mitigate the
propagating effects of counterparties’ defaults on their respective obligations.94 While the logic
of this requirement is compelling, it proved to be the cause of a different set of vulnerabilities.
It has been observed, in light of both Archegos and the LDI crisis, that the use of collateral and
88
Regulation (EU) No 648/2012 of the European Parliament and of the Council of 4 July 2012.
89
Ibid, art.4. It needs to be noted that CCPs are authorised by the relevant national regulator, so ESMA in the EU
regulatory framework, and now the FCA in the UK.
90
Regulation (EU) No 648/2012, art.2.
91
Dodd-Frank Wall Street Reform and Consumer Protection Act 2010, Pub. L. 111–203 S.731.
92
See ESMA “The Clearing Obligation under EMIR”, Discussion Paper 12 July 2013, 2013/925; see also See
Dodd Frank Act, s. 723. The definition of which types of swaps and which counterparties should be exempted
from clearing requirements has been made by the SEC and the Commodity Futures Trading Commission.
93
Regulation (EU) No 648/2012, art. 41.
94
Ibid. Margins are in essence a way for a CCP to collateralise its exposure to clearing members, and they are
based on the calculation of the risk characteristics of the products cleared.
18
margin calls has the potential to create liquidity spirals. The mechanics of margin requirements
imply that changes in the price of underlying assets trigger higher initial margins due to the
increased volatility in the asset market, and critically, larger margin calls requested by
counterparties to meet current losses throughout the life of the contract.95
In the context of the LDI crisis, margin calls forced counterparties in swap contracts to
access credit facilities in order to cover collateral needs. These sudden margin calls were of
course caused by sharp moves in interest rates (and in turn in the gilt market), and most
investors were unprepared to withstand the resulting calls for cash. Counterparties unable to
find collateral to cover their position were either facing the prospect of defaulting, or had to
resort to asset sales to find the necessary cash, causing in the process further shocks in the
underlying asset market (the gilt market in the context of the LDI crisis).96
These dynamics of liquidity spirals are also a source of systemic vulnerability, because
margin calls, by effectively triggering either defaults or fire-sales, perpetuate the further decline
in the price of the underlying asset, creating in the process layers of interconnectedness among
different entities involved in the derivatives market. It is also worth noting that problems
associated with margin calls are not new, and in fact follow pro-cyclical patterns that were
clearly identified in past crises, and also in the context of more general leverage cycles.97
What becomes increasingly evident in the context of the LDI crisis is that the role
traditionally ascribed to derivatives, of instruments of risk mitigation and risk management,
shifted over the past decades to encompass a purely speculative function. From the perspective
of investors looking to hedge specific exposures (or in the case of pension funds, trying to
hedge long-term liabilities), margin requirement often result in risks being magnified if the
swap is out of the money and there is an increase in market volatility. In particular, as was
explained in the previous section, the impact of volatility on derivatives’ counterparties is
amplified by the level of leverage that they trade with.
Problems of volatility and leverage were compounded by the increased complexity of
transactions, and the quantitative models underscoring their rationale. The level of leverage
employed by some pension funds led commentators to argue that it was chiefly incentive-
95
In the Archegos case for instance the decline in the value of the stock that Archegos was exposed to via TRSs
triggered the request of huge variation margins by Archegos’ counterparties. Supra ESMA, May 2022, p.8. See
also Satyajit Das “The risks from derivatives have morphed”, Financial Times, October 7, 2022.
96
Supra Das, Financial Times 2022; and ESMA, 2022, p.8.
97
John Geanakoplos for instance observed that the pro-cyclicality is the key element of these rapid swings and
associated calls for cash, which initially affect the most leveraged counterparties, but eventually affect a wider
spectrum of market participants with prices spiralling down, and reinforcing a dynamic that leads to a crisis
situation. See John Geanakoplos “Solving the present crisis and managing the leverage cycle”, Economic Policy
Review - Federal Reserve Bank of New Yor 2010, p.104.
19
based: not really aimed at creating a hedge, but simply at profiting, because investment advisors
were being compensated for the complexity of their advice.98 While the central thrust of this
enquiry is not on the governance of pension scheme and on the role of advisors and asset
managers, questions related to the suitability and appropriateness of these products resurface,
bearing similarities with recent mis-selling scandals involving the same products.99 In
particular, in the context of the LDI crisis, the complexity of the swaps, and the calculations of
variation margins that could be triggered by changes in interest rates, made it very difficult (if
not impossible) for pension schemes, and their trustees, to assess the consequences of worst
case scenarios and wrong way risks among counterparties.100 Some attendant concerns are
further explored in the next section.
98
See John Rolfe “Investigation needed to hold those behind UK pension crisis to account”, Financial Times,
2022. John Rolfe, a pioneer himself of LDI strategies at Boots, voiced repeatedly his concern about the use of
leverage in LDIs arguing that it amounted to pure and unnecessary speculation. He also compared pension
schemes to poorly run hedge funds.
99
See on this Vincenzo Bavoso “Financial Innovation, Derivatives and the UK and US Interest Rate Swap
Scandals: Drawing New Boundaries for the Regulation of Financial Innovation”, Global Policy Vol.7 Issue 2,
2016.
100
See Harriet Agnew, Adrienne Klasa, Josephine Cumbo, Chris Flood, and Anjli Raval “How bond market
mayhem set off a pension “time bomb””, Financial Times, October 8 2022.
101
See Sirio Aramonte, Andreas Schrimpf and Hyun Song Shin “Non-bank financial intermediaries and financial
stability”, BIS Working Paper N.972, October 2021. See also BIS “Newsletter on bank exposures to non-bank
financial intermediaries”, 24 November 2022.
102
On this see for a discussion Vincenzo Bavoso “Basel III and the Regulation of Market-Based Finance: The
Tentative Reform”, NYU Journal of Law and Business, Vol.18 N.1, 2021.
103
Supra Aramonte, Schrimpf and Shin, 2021, highlighting the central role of leverage fluctuations and the
changes in margin requirements.
20
104
See for instance relevant provisions within Basel III, which, inter alia, introduced new frameworks for leverage
and liquidity. See for an analysis supra Bavoso 2021.
105
Such as for instance the leverage ratio, the liquidity covered ratio, and the net stable funding ratio, all introduced
under Basel III. See for an overview BIS “Basel III: international regulatory framework for banks”, available at
https://www.bis.org/bcbs/basel3.htm.
106
Financial Stability Board “Implementation and Effects of the G20 Financial Regulatory Reforms”, 5th Annual
Report, 2019, p.22-26; see also BIS “Newsletter on bank exposures to non-bank financial intermediaries”, 24
November 2022, where it is specifically observed that systemic vulnerabilities arise due to banks’ exposures to
channels of non-bank intermediation, and particularly to highly leveraged counterparties due to derivatives and
securities financing (repo facilities).
107
Bearing in mind that the Basel III framework did introduce a leverage ratio that applies to banking institutions,
supra fn.77.
108
See Speech by Sarah Breeden “Risks from leverage: How did a small corner of the pension industry threaten
financial stability?”, Bank of England, 7 November 2022, p.4. It should be noted that a rather different view of
leverage, and particularly of private debt creation, was provided by Adair Turner in his book “Between Debt and
the Devil”, Princeton University Press, 2015, where Turner observed how most credit created through the banking
system is not directed at economic growth, but it is instead invested towards already existing assets (such as
housing), thus leading ultimately to asset price bubbles.
109
Supra Breeden 2022.
21
With respect to the second point, Sarah Breeden pointed in her speech at clearing and margin
requirements, just as was seen in the previous section, as the factors that while mitigating
counterparty credit risks, increased mechanisms of systemic risk transmission.110 Drawing on
the case of Archegos, she also highlighted the information problem that swap counterparties
have in assessing each other’s risks – arguing that margin requirements can be effective only
where counterparty credit risk can be properly assessed by the counterparties.111 Of course,
questions of transparency and imperfect information become only more problematic in the
context of synthetic leverage acquired though derivative positions.
In a similar vein, Professor Steven Schwarcz recently criticised the impact of post-2008
regulation on derivatives, and again his main focus was the introduction, tout court, of clearing
and margin requirements.112 While providing a very useful reconceptualisation of derivatives
along two contractual families, namely options and guarantees, and attributing to the latter (and
specifically to credit default swaps113) the greatest capacity to create systemic risks and
propagate them, Professor Schwarcz probably downplays the potential that all derivatives have
to create speculative leverage.114
It has been correctly observed by a number of scholars and commentators 115 that the
interposition of CCPs in derivatives transactions has effectively concentrated risks in central
clearing organisations, creating in the process some degree of moral hazard, because, as
Schwarcz argues, CCPs ensure the performance of a derivative contract even where the
counterparty fails on its obligation.116 These concerns however, while being well conceived,
sit oddly with the further suggestion that outside of the banking sector the main tool of risk-
reduction and resilience-building should be placed within the individual firms and their
capacity to manage their use of leverage.117
110
She pointed at the same problem as the root of systemic risks both within the framework of market channel of
risk and counterparty channel of risk. Ibid, p.6-7.
111
Ibid, p.7. This was clearly one of the problems that emerged in the collapse of Archegos in the US. See ESMA
“Leverage and derivatives – the case of Archegos”, May 2022.
112
See Steven Schwarcz “Post-Crisis Derivatives Regulation: What Went Right (and What Went Wrong)”,
Science Po Law Review, January 2023 (Biennial issue on Financial Regulation after the 2008 Crisis).
113
Ibid, p.114-115, due to their capacity to create interconnectedness and correlation of risks.
114
See in this sense Mark Loewenstein “Speculation and Leverage”, Asian Finance Association Conference, 2013,
available at SSRN http://ssrn.com/abstract=2269889, where he argues that speculative episodes take a problematic
turn when they are accompanied by highly leveraged positions.
115
See on this the work of Yesha Yadav “The Problematic Case of Clearinghouses in Complex Markets”, 101
Georgetown Law Journal 387, 2013.
116
Supra Schwarcz 2023, p.124.
117
In this sense supra Breeden 2022, p.8. In a similar way Schwarcz argued that much of post-2008 derivatives
regulation results in over-regulating sophisticated firms that are capable of assessing and managing their own risk
exposures. Supra Schwarcz 2023, p.127-128.
22
118
See Margaret Blair “Financial Innovation, Leverage, Bubbles, and the Distribution of Income”, 225 Review of
Banking & Financial Law, Vol.30, 2010-11.
119
See on this Emilios Avgouleas and Jay Cullen “Excessive Leverage and Bankers’ Pay: Governance and
Financial Stability Costs of a Symbiotic Relationship”, Columbia Journal of European Law, Vol.21 N.1, 2015.
The authors provide an exhaustive argument showing the incentives within financial institutions to engage with
high levels of leverage and the cognitive bias (such as herding) that prevent individual firms to act rationally and
limit their risk-taking and leverage. They argue in other words in favour of an overarching regulation of leverage.
120
See Emilios Avgouleas “The Global Financial Crisis and the Disclosure Paradigm in European Financial
Regulation: The Case for Reform”, European Company and Financial Law Review, Vol 4, 2009, p.453.
121
Saule Omarova “From Reaction to Prevention: Product Approval as a Model of Derivatives Regulation”,
Harvard Business Law Review Online, Vol.3, 2013, p.99. Omarova argues, in the US context, that the Dodd-
Frank failed to impose direct and targeted regulatory constraints on market actors, and instead it resorted to the
old technique of regulating by enhancing information flows.
122
See in this sense John Rolfe “Investigation needed to hold those behind UK pension crisis to account”,
Financial Times, 2022; John Plender “Lessons from the Gilt Crisis”, 21 December 2022. See also on the question
of speculation Hanif Virji “Capacity: Is the Question of Hedging or Speculation Mis-stated?”, Butterworth Journal
of International Banking and Financial Law, 602 October 2022.
123
Supra Rolfe 2022; and Plender 2022.
124
Rodney Sullivan “Speculative Leverage: A False Cure for Pension Woes”, Financial Analysis Journal, 66:3,
2010, p.6. Sullivan mentions in particular “the deceptive allure of leverage … to meet otherwise unachievable
annual total return targets”. Similarly, and in a wider context, Steve Eisman commenting on the global financial
crisis of 2008, indicated leverage as one of the main causes, stating that bank executives “mistook leverage for
23
pension schemes is a legitimate strategy, the higher risk that leverage entails should not be
underestimated, particularly given the probability of negative events being higher than
normal.125 It has been further argued in finance that, while a liability-matched portfolio may
indeed require some degree of leverage, this leverage should be non-speculative, namely it
should be used to hedge and/or remove risks, not add them.126
Ultimately, the employment of leverage in pension schemes is also the result of the
application of so-called risk parity in the context of portfolio management. Some commentators
have seen the move from more traditional strategies (such as the employment of Modern
Portfolio Theory127) to risk parity as a way to attribute theoretical legitimisation to the use of
leverage, and with it the employment of much more complex mathematical modelling, given
the sophistication of the attendant finance theory.128 This last reflection confirms what was
suggested earlier in this section, that the use of leverage (or at least the excessive use of it)
through complex derivatives was the result of a system of culture and incentives that has come
to permeate financial markets over the past decades, and that has remained intact even after the
global financial crisis of 2008.129
In light of the disputed role of post-2008 financial regulation in curbing excesses in
derivatives markets, this enquiry turns to the potential that legal doctrines have to achieve that
goal.
genius”. See Steve Eisman | Wall Street Debate | Opposition (4/8) available at
https://www.youtube.com/watch?v=73OZncDEDks.
125
Supra Sullivan 2010, p.6. In finance this is referred to as “fat tails”, which essentially is the greater than
expected probability of extreme values.
126
Ibid, p.8.
127
This being the classic work on risk diversification, it was pioneered by Harry Markovitz “Portfolio Selection”,
Journal of Finance, Vol.7 N.1, 1952.
128
See Laurence Siegel “Risk Parity: Classical Finance Properly Implemented, or Misunderstood?”, Financial
Analysts Journal, September/October 2010, Vol.66, N.5, p.15.
129
See for a discussion Ismail Erturk, Julie Froud, Sukhdev Johal, Adam Leaver and Karel Williams “The
Democratization of Finance? Promises, Outcomes and Conditions”, Review of International Political Economy,
14:4, October 2007.
130
Supra Omarova 2013, p.99.
24
more effective limitation of speculative leveraged exposures can be achieved by looking at the
legal concepts explored in section two, particularly the way in which they can incorporate a
proprietary element into derivatives.
A return to a tout court contractual requirement that mandate the deliverability of the
underlying physical assets is admittedly difficult to conceive. In his work on the evolution of
derivatives, Donald MacKenzie documented the progressive “virtualisation” of derivatives,
which became far removed from the assets and markets underscoring them.131 The physical
deliverability of assets would in other words be impeded by two factors. First, the widespread
application of speculative leverage has created a derivatives market whose value is much
greater than the value of the financial assets that in principle reference derivatives contracts132
- therefore there would be a scarcity of assets if physical delivery were to be re-introduced.
Second, derivatives have progressively moved away from commodities and agricultural
markets whose value they were historically hedging, only to embrace from the 1980s onwards
financial assets and indexes as main reference assets.133 These financial assets only exist
presently in dematerialised form, which means that the configuration of their delivery is hard
to fathom. Despite all this, the concept of deliverability is important because of the proprietary
interest it creates, and legislative efforts could be directed at devising and re-interpreting a
deliverability criterion that could be employed in present-day financial markets.
Undoubtedly, the progressive ease of the deliverability requirement gave way to more
cost-effective contractual structures, so much so that cash settlement became the standard
contractual form and derivatives are commonly referred to as contracts for differences in most
statutory texts. This further depletes derivatives of any sufficient proprietary interest in the
underlying assets, and of course plays against any substantial control of the speculative
leverage that is created in the process. When it comes to contractual standardisation though,
given current market structures, the role of ISDA (a private association of market participants)
is central to the standardisation of contractual terms. This is exemplified for instance by its
provision of different methods of cash settlements.134 As a result, changes to the contractual
form of derivatives would necessarily hinge on ISDA’s willingness to do so, which in turn
131
See Donald MacKenzie “Material Markets: How Economic Agents are Constructed”, Oxford University Press
2009, chapter 4, discussing in particular derivatives as the product of virtuality.
132
See for instance Jacob Leibenluft “$596 Trillion! How can the derivatives market be worth more than the
world’s total financial assets?”, Slate, 15 October 2008, where it was observed that the value of global derivatives
markets was over five times bigger than the value of the world’s financial assets.
133
See Donald MacKenzie “An Engine, Not a Camera: How Financial Models Shape Markets”, MIT Press 2008,
chapter 2.
134
See ISDA “Cash Settlement Methods in the 2021 Definitions”, available at https://www.isda.org/a/ecigE/Cash-
Settlement-Methods-cheat-sheet.pdf.
25
6 – Concluding remarks
The crisis in the UK LDI schemes shows, above everything else, the interconnections
between the pension system and global finance. This is the result of policies that have been
directed, for over thirty years, at promoting financial reforms that advanced precisely the
135
See for a detailed analysis Gary Meggitt “Insurable Interest – The Doctrine that Would not Die”, Legal Studies,
Vol.35 N.2, 2015, p.287.
136
Opinion prepared for the ISDA by Robin Potts QC, Erskine Chambers, 24 June 1997, para 5.
137
The use of credit default swaps in the pre-2008 years is emblematic of this practice, which was spectacularly
epitomised by the short positions secretly entered into by hedge funds in trades such as the Abacus CDO or the
Magnetar Trade. See for a discussion Vincenzo Bavoso “Filling the Accountability Gap in Structured Finance
Transactions: The Case for a Broader Fiduciary Obligation”, Columbia Journal of European Law, Vol.23, N.2,
2016.
138
See Oskari Juurikkala “Credit Default Swaps and Insurance: Against the Potts Opinion”, Journal of
International Banking Law and Regulation, Volume 26, Issue 3, 2011.
26
dependence of the pension system on financial markets – the accounting reforms signposted at
the beginning of section two exemplify this process. This shift was accompanied, as was also
discussed in section two, by a progressive liberalisation of the financial system, whose
development and growth took priority over the stability of the pension system and the welfare
of its members.139 As was explained throughout this article, UK pension schemes, and
particularly LDIs, have become an important cog in financial markets dynamics, trapped in
(but also contributing to) leverage and liquidity dynamics.
With the Bank of England called upon the task of stabilising markets – both the gilt
market and pension schemes – it became more evident that 1) risks and fragilities in this area
of the financial system had acquired a systemic dimension, and 2) the same risks required
mutualisation through the Bank of England’s intervention.
This turn of events raises a number of questions, which this article addressed. Starting
with the macro-problem that emerged in the context of the LDI crisis, it is becoming ever more
apparent that risks of a systemic nature originate in peripheral segments of the financial system.
While these segments, such as the pension system, were not the main focus of regulatory
attention in the post-2008 years, they have demonstrated their capacity to propagate risks
through leverage creation and liquidity spirals. In this sense, the LDI crisis replicated similar
instances of financial instability that in recent years emanated from the non-bank sector.
From a micro-angle of analysis, some of the risks, and importantly their transmission
across the financial system, became possible due to the use of derivatives. This article focused
specifically on the employment of interest rate swaps and total return swaps in LDIs, and how
their use elicited high levels of leverage. Moreover, due to the interconnectedness that is an
intrinsic feature of credit derivatives, leverage and liquidity risks quickly spread across the
financial system, investing more critically also large financial institutions that were indirectly
exposed to said risks.140
While critics of post-2008 regulation in the area of derivatives have correctly pointed
at the problematic role of CCPs as concentrations of mutualised risks, potentially contributing
to moral hazard, CCPs remain an important piece in the framework designed to mitigate
systemic risks. More regulatory question though should have been asked, further to this latest
episode, about the misuse of derivatives and the excessive leverage that they elicited,
139
This is part of a broader debate on the role of the City of London (and Wall Street in the US) in the context of
wider UK policies and economic goals, whereby it is observed that the interests of the City are made to coincide
with wider social and economic interests. See the edited collection “Integrity, Risk and Accountability in Capital
Markets: Regulating Culture”, Edited by Justin O’Brien and George Gilligan, Hart, 2013, particularly part I.
140
Supra Financial Stability Board 2019.
27
particularly as this happened in the non-bank sector, outside the umbrella of leverage and
liquidity regulation. Why, in other words, are non-bank entities allowed to engage with forms
of intermediation that are intrinsically risky, and create leverage and liquidity transformation
without the necessary regulatory backstops? Post-2008 regulation, both in Europe and in the
US, has not addressed this fundamental regulatory question, namely the amount of risk that
derivatives markets can create beyond what becomes socially desirable.141
It is precisely in response to this question that this article explored a number of legal
doctrines, whose re-deployment could contribute to the control of speculative leverage. This
article highlighted how the element of deliverability (its partial reintroduction), the contractual
structure revolving around cash settlement (limiting its availability), and the insurable interest
(its customised application) could all perform a useful function in reining in the risks that
emanate from derivatives market. While these changes are not readily available, and would
require some form of legislative intervention (at least in the UK and the US which are the main
venues of derivatives trading), there is a body of positive law that could be drawn upon to
reinterpret these doctrines and adapt them to current circumstances.
As was explained in section two of this article, derivatives were traditionally limited by
the application of these legal principles (particularly the first two), and the rationale for their
re-introduction rests in the need to preserve their hedging function, while limiting their capacity
to create speculation, leverage, and the transmission of systemic risk. The LDI crisis shows
exactly this: how a good idea – using derivatives to hedge interest rate and inflation risks –
became the perfect tool to inflate profits (for the much expanded chain of consultants and asset
managers) and spread undue risks through the excessive use of leverage.
Statement:
1) there are no conflicts of interest, 2) there is no funding to be declared for this paper, 3) there
is no dataset that was used to complete this paper.
141
In the context of the LDI crisis, the social costs of excessive risk-taking through derivatives was mutualised
through the central bank’s intervention. On different ways to regulate derivatives in order to limit their socially
harmful consequences, see supra Omarova, 2013, who proposed a product approval system.
28