Blog

"It's not just whom you sleep with, but also whom they are sleeping with," Warren Buffett told investors back in 2008. He was referring to the chain of connections you are exposed to when dealing with a derivatives counterparty. His biological analogy, familiar to those studying the spread of viruses, was picked up in a speech by the Bank of England’s Andrew Haldane the following year.

The financial system resembles a “cat’s cradle of interconnections”, said Haldane. It’s an adaptive, complex network that is prone to seizing up.

Identifying and analysing the problem has proved much easier than dealing with it. One solution highlighted by Haldane in his paper—broadening the network of central counterparties and requiring intra-system netting arrangements—is a major feature of recent regulatory reform (via the US Dodd-Frank Act and its equivalent in Europe, the European Market Infrastructure Regulation).

But proposed changes in derivatives regulation have met with fierce lobbying by affected financial institutions. Banks and other dealers are reluctant to give up the fat profit margins generated by dealing on a bilateral basis and they don’t want to suffer the cost of the strict collateral requirements that a central counterparty would impose. Buffett, controversially, has been among the lobbyists, meriting accusations of hypocrisy.

Another solution to network complexity—what Haldane called “seeking actively to vaccinate the ‘super-spreaders’ of financial contagion”—is turning out to be even more difficult to deliver.

On the one hand regulators are asking so-called “systemically important financial institutions” to hold higher levels of capital than those banks that do not act as such vital nodes in the global system. “Living wills” are supposed to enable such large institutions to be wound down more painlessly in the event of failure.

At the same time regulators are under pressure to avoid too rapid a deleveraging of the financial system. Politicians and central bank heads are desperate to avoid a sharp recession. Apart from threatening incumbent governments with the likely loss of power, a recession will certainly produce a surge in debt defaults, with sovereign issuers’ creditworthiness now in major doubt too, unlike in 2008. Such considerations are undoubtedly behind recent calls from the Bank of England to ease lending rules, even if this contradicts much of the tough talk issued post-crisis.

Unfortunately, these mixed messages imply that financial market reform efforts may be heading towards paralysis. Meanwhile, markets are well on the way to forcing the issue.

As Jonathan Weil of Bloomberg pointed out yesterday, investors are already valuing a number of European banks at levels well below the net asset figures recorded on their balance sheets. In other words, those assets are being incorrectly valued, or are deteriorating in value, or both. The stock market value of UniCredit, which announced a third-quarter loss of €10.6 billion earlier this week, is only 28 percent of the bank’s book value, for example. In the US, Bank of America now trades at a very similar ratio, its stock price well south of the point at which Buffett’s Berkshire Hathaway fund made a US$5 billion preferred equity investment in late August.

These stock market valuations of banks tell us to expect a major acceleration of announcements of losses and loan writedowns. Capital will have to be raised in a hurry, while shrinking equity bases at banks are forcing them to reduce assets at an even faster rate. Banks’ loan books are shrinking, however hard politicians and regulators are trying to prop them up.

Meanwhile, it’s no surprise that institutional investors are reportedly scrambling to work out where and to whom they are carrying exposures, where their assets are being held in custody and whether they can access collateral in an emergency. The recent failure of MF Global and the ensuing tussle over missing and supposedly segregated client assets have only served to heighten counterparty concerns.

System-wide gearing and high levels of interconnectedness provide a tailwind for everyone during a debt-fuelled boom, but they are a curse when the cycle reverses, as historians of the 1930s could have told us. You can’t get out of debt without shrinking your earnings, which exacerbates the problem. And there aren’t enough truly safe havens for cash within the financial system, as Robert Dubois recently highlighted.

For various reasons, a significant part of the intensifying debate over counterparty and collateral risks in financial products has centred on ETFs. Many ETF industry insiders complain that they are being unfairly singled out for criticism, when the same or even greater risks apply in other products too.

How risks in tracker funds should be assessed in a broader context is the key theme of our upcoming IndexUniverse.eu webinar, to be held on December 1. Are you really at risk of financial market contagion when buying an ETF? How safe are different types of fund, and what about other funds and bank-issued investment vehicles? I’ll be joined by three expert panelists to discuss these important issues, and hope you will be able to join us.

Meanwhile, with governments now heading towards defaults, the threat of contagion in the broader financial markets is at its highest level for over 70 years. Vaccination schemes have come too little, too late, and the transmission mechanisms are too large and complex to be easily dismantled.

As a rallying cry, “asset managers are taking small and well-controlled risks to achieve solid but unexciting rewards” doesn't have the same ring as “put the cat back in the bag” or, dare I say, ”the sky is falling!” For passive investment managers, hyperbole comes hard. It just doesn’t feature in the job description. So we will leave the headline writing to others, but it doesn't mean we don't feel strongly. Indeed, in our view the current “debate” around ETFs is overweight conclusions and underweight facts and analysis. Here are a few thoughts that attempt to redress the imbalance.

First, exchange-traded funds work. The ETF industry has an unparalleled record in delivering on its performance promises: efficient benchmark tracking, relatively low cost, high transparency and improved liquidity when compared with conventional funds. For more than 20 years, these products have delivered on their promises. This is equally true, regardless of whether the funds are physical or synthetic, or whether they engage in securities lending or not. (I would concede that some optimised physical funds have struggled with tracking error on occasion, but there’s nothing here that would warrant the deep and sustained attention of regulators or the media). In particular, the approach of the ETF industry to transparency puts most other investment products to shame.

Second, taking counterparty risk is not dumb. ETFs take relatively small risks in order to improve the effectiveness of their benchmark tracking. Some funds engage in securities lending to generate supplemental revenue to make up for other sources of tracking error. So-called “synthetic” ETFs use swap contracts to lock in a specific level of benchmark performance. Both types of fund rely on counterparty performance. Both use broadly accepted and effective risk mitigation techniques to reduce the potential for loss. Indeed, the ETF industry has used these techniques since its inception with increasing effectiveness and, to our knowledge, no material adverse consequence.

A useful analogy is property ownership. If you purchase a property for investment you can choose whether to rent it out or leave it empty. If you take on a tenant, you incur risk: they may not pay the rent, they might stain the carpet, they might invite the local chapter of “squatters anonymous” to stay for the weekend. You can avoid these risks by keeping the property empty. But then you forgo the rent. You can mitigate your risk by conducting background checks, insisting on references and a salary history and inspecting the property on a regular basis, but it will not go away completely. Is it inherently stupid to become a landlord? No. But neither is it simple. However, the simple approach (no tenants, no rent) is unlikely to be the wise choice.

Third, life is relative. ETF providers did not invent derivatives, counterparty risk, securities lending or even “swap-backed” funds. Indeed, in Europe there are €6 trillion of assets in UCITS funds and nearly half of these assets (approximately €2.8 trillion) are currently available for securities lending. By contrast, the assets under management of the entire European ETF industry are less than €250 billion. In other words, there are ten times more assets sitting in UCITS funds that are available for securities lending than are managed by the whole of the ETF industry!

So why is this debate about ETFs? Perhaps it is because they are transparent. Perhaps it is because ETF issuers talk to their clients (endlessly and boringly) about their structures. Perhaps it is because we criticise one another and compete on the basis of continuously improving risk management techniques. Regardless, while it is important and appropriate that investors understand how ETFs operate, it is also important and appropriate that they have a comparable understanding of their investment alternatives.

On any meaningful measure (cost, performance, transparency, liquidity, structural risk), any objective evaluation would award high marks to ETFs. Compared to conventional funds, ETFs offer greater transparency, a higher certainty of performance, improved liquidity, lower fees and, typically, comparable levels of counterparty risk. Compared to structured notes, they offer substantially reduced counterparty risk, simplicity of investment thesis and lower fees. Compared to cash accounts and government bonds, they offer the potential for future appreciation in a world of historically low interest rates.

Investing is not easy. Complexity and safety are not antithetical. Nor are simplicity and safety synonymous.

Ted Hood is chief executive officer of Source, a European ETF provider

A recent online survey of independent financial advisers (IFAs) in the UK found that 60 percent would not recommend a swap-based ETF. The feedback given by respondents suggested that this was in response to various regulators and other pan-European institutions asking some very pointed questions about ETFs—especially the derivative-based (or “synthetic”) variants of these increasingly high-profile vehicles.

This is a worry, as almost all European ETFs are constituted and regulated as UCITS funds, and UCITS are supposed to be a product for the average man or woman on the average European street.

The issue of counterparty risk associated with swaps is the core concern of both IFAs and the regulators. IFAs will remember that the UK regulator issued some very strong words (and fines) when counterparty risk was last raised in connection with “guaranteed” or “structured” products.

Collateral or collateral?

Possibly nudged on by active managers, who are keen to see low-cost ETFs get “a good kicking”, the regulators then started asking tricky questions about securities lending too. Is the economic exposure created by securities lending the same as in swap-based ETFs using reference baskets? Both have collateral posted, and this collateral may differ in quality and behaviour to the index being tracked.

The securities lending lobby will point to governance requirements such as the standards set by the FSA in the UK, where market participants have drawn up the Securities Borrowing and Lending Code of Guidance (2009). It was issued by the Securities Lending and Repo Committee, a committee that is chaired and administered by the Bank of England and is observed as a guide to good practice.

Additionally, the Global Master Securities Lending Agreement (GMSLA) is a best practice agreement drafted by the International Securities Lending Association (ISLA), which is used by many participants.

The swap-based fan club will refer to the fact that collateral held to offset derivatives exposure must comply with Committee of European Securities Regulators guidelines (CESR/10-788). It is also true that an increasing number of synthetic issuers allow you to see the basket holdings and exposure daily, whereas with securities lending there is no equivalent disclosure.

The regulators should be encouraged that all ETF players have responded so quickly with greater transparency, even if the players don’t all agree on whether physical or derivative-based structures are best.

Another concern is that swap-based ETF providers suggest that the possible loss resulting from counterparty exposure is limited to 10 percent (the level at which the exposure to a counterparty under a swap is subject to a “reset”), when in fact the losses could be higher in the event of a default by the swap provider (as a result of “gap risk”, a shortfall when collateral assets are sold or as the result of a delay in gaining access to collateral).

There will be concerns over the collateral received by funds in securities lending operations even though this typically exceeds the value of stock or bonds lent and there are limits on the amount of the fund lent. Some lending agents provide indemnification to the fund, but further disclosure in this area would be welcomed by professional investors. What is the exposure to each borrower, for example?

The expectation of collateral is that it will be there, in sufficient quantity and in liquid form, when you most need it. This may not always be the case.

But the risk is everywhere

The regulatory spotlight has so far only been shone on ETFs. But there are many other occasions when retail investors might be exposed to swaps, derivatives or lending risks.

We can compare index mutual funds for starters—there is a similar level of securities lending risk here—yet do index funds have the same level of disclosure as, for example, iShares ETFs?

Where else might large scale securities lending be taking place? Active funds? Life funds? Insurance based pension funds? Yes to all of those. Is the process and collateralisation transparent? And is the fee share disclosed?

Where swaps and other derivatives are involved, what is the level of disclosure of risk within non-index UCITS funds? And what about life funds—do they not often use OTC derivatives to protect solvency? Are these well collateralised, well priced and transparent? Who knows?

Don’t be ridiculous, you might say. Of course we can’t have full disclosure of every derivative instrument in a UCITS or any other retail fund. But the issue is that investors are supposed to be reassured when a fund is compliant with the UCITS rules and, in my opinion, these rules have been flexed to allow inappropriate and complicated structures such as synthetic ETFs with mismatched reference baskets inside the regime.

What next?

The UCITS regime was originally intended to create simple, diversified products that could be sold in high streets across Europe. UCITS III and IV have introduced levels of complexity that didn’t exist before. While the objective may be to reduce risk, that will only be achieved if there is a clear and honest demonstration of aligned interests, together with a full disclosure of costs, revenues and counterparties, that the man or woman in the high street can trust. Anything less increases the risk to one of the greatest ever investment brands, UCITS.

I’ve often heard ETF issuers say that as long as their fund does what it’s supposed to do and tracks its index, who cares how it’s put together? Investors should consider their ETF as a car or a mobile phone, issuers suggest, and value a fund for its utility, not its engineering.

This assertion, however, ignores two key problems.

First, someone in search of a new phone still needs to choose between the tens of thousands of different deals on offer.

Do you want pay-as-you-go, a 12-, 24- or 36- month contract or a SIM-only deal? Would you prefer your iPhone 4 with 8, 16 or 32 gigabytes of memory, do you want to pay more for the “S” at the end, how many “free” minutes of calling do you need, and how much internet access do you want to get charged for in addition?

At this point I suspect that 99 percent of us give up. And, unfortunately, there are no easy-to-understand comparisons available to guide us on our way by reducing product comparisons to simple, like-for-like measures.

Of course, if there are several competing providers of services, you might assume that the cost of any contract is going to bear some resemblance to its real worth. So perhaps it’s not necessary to spend too much time tearing your hair out worrying about comparability. You may not arrive at the absolute best deal, but you should be thereabouts.

But is that a fair assumption? One mobile phone operator may undercut competitors, but if you sign up with the company you may discover that its network coverage is poor, you can’t get a signal at home and you’ve just locked yourself into a three-year contract.

The Bank of England suggested this summer that the lower-than-average total expense ratios charged by European synthetic ETFs, when compared with those levied by funds using physical replication, might reflect synergies between bank-owned asset managers and their parent banks’ trading desks. But those lower charges could also disguise inbuilt risks that are not being fully understood by investors, the UK’s central bank theorised.

To give another example, iShares recently told the market that to collateralise its Hong Kong-based FTSE A50 China Index ETF would cost 50-100 basis points on top of the fund’s existing total expense ratio of 1.39 percent, taking costs to over 2 percent a year. But Deutsche Bank, via its Singapore-listed CSI 300 Index ETF, is offering collateralised exposure to Chinese A shares (albeit via a different index), for 50 basis points, a quarter of the cost of the iShares fund.

Why the difference? Is db x-trackers subsidising its fees for this particular fund to attract assets from the iShares ETF, the largest tracker fund in the region? Is the bank making up the gap in fees via securities lending, trading, or a difference in the way it provides collateral backing? If iShares is really being charged 100 basis points by third parties for collateralised exposure to A shares, why doesn’t Deutsche Bank sell its quota to iShares, rather than to its in-house ETF? These are all fair questions to ask. On the basis of publicly available information, it’s impossible to answer them.

This brings us to the second problem, which also goes to the heart of the financial crisis.

During last Friday’s stimulating debate at the Centre for European Policy Studies in Brussels on the ETF market and UCITS, one issuer made the standard comparison between ETFs and cars, and argued that if your investment vehicle gets you where you want to go, do you really need to worry about how its engine works?

The regulators at the CEPS event responded that the simple act of driving a car imposes direct costs (in economists’ jargon, “negative externalities”) on others: both on other drivers (by adding to congestion) and on the general public (via pollution). Public policymakers therefore need to step in and set some ground rules, regulators concluded, surely indisputably.

It’s not difficult to see how the financial industry has externalised its own costs, leaving taxpayers to pick up the bill for the aftermath of the credit bubble (and the bills for several bailouts over previous decades, too).

Unfortunately, global financial regulators seem to be heading in precisely the wrong direction, enshrining “too big to fail” status at so-called “systemically important financial institutions” (albeit with higher capital standards for those on the list) rather than charting a path towards the total removal of government support for banks.

At Friday’s ETF-focused debate in Brussels, it was clear that regulators were particularly concerned about the possible risks arising from synthetic ETFs precisely because many of the banks issuing them have such systemic importance. Those among the bank-backed ETF issuers who complain that they are not being treated equally with issuers of physically backed funds, typically asset management firms (for example in the assessment of collateral exposures), surely have to accept this point.

But systemic risk contagion could arise from the asset management sector, too. Take the US$2.6 trillion US money market fund (MMF) business, where the promise of a minimum US$1 per share redemption value, enshrined by convention and hinted at in marketing literature, is in fact not a guarantee at all. Meanwhile, maintaining a stable NAV under conditions of near-zero interest rates is becoming tougher and tougher.

As Robert Dubois has ably described on IndexUniverse.eu, there’s still limited transparency into how exactly these funds operate. While the US government stepped in and guaranteed money market fund NAVs at US$1 in 2008, there’s presumably less likelihood that it would be able to do so in our current, straitened times. But even though it seems like an obvious solution just to tell investors that they may not receive a minimum US$1 per share back for their MMF holdings, the US SEC is still reluctant to accept the need for a fully floating (i.e. market-driven) NAV for these funds. Why? The regulators are panicked at the prospect of a run on MMFs.

How all these competing forces can be reconciled and whether regulators can ever arrive at a level playing field is debatable. On Friday in Brussels, the talk was of undertaking a broader review of the UCITS rules and a deeper examination of shadow banking—repo market, money market funds, securities lending and all. All this smacks of a discussion that could extend several years into the future. That’s good news for regulators, lawyers and journalists like me, who will get to report on it. But not, I suspect, for consumers of financial products, nor for the world economy.

Whether markets will wait that long (and not force government and bank defaults long before) is a very good question.

Meanwhile, as if they didn’t have other things to worry about, investors now need to keep an ear open for complex regulatory discussions when selecting a financial product and product provider. Getting from A to B sounds simple, but travellers may face unexpected risks, unmapped detours and a sudden imposition of tolls along the road.

In the film Casablanca, Rick (Humphrey Bogart) tells Captain Renault (Claude Rains) that he came to the city because of his health, for the waters. Informed that they are in the desert, Rick replies, with an ironic tone, that he has been ”misinformed”. Investors and banks that purchased Greek sovereign credit default swaps (CDS) to protect themselves against the risk of the country’s default may find that they have been similarly “misinformed”.

As part of the “grand plan” to resolve Europe’s ongoing debt crisis, the Institute of International Finance (IIF), a global trade body representing bankers, voluntarily accepted a 50 percent “haircut” on its holdings of Greek bonds. The associated linguistic gymnastics, with default redefined as “restructuring”, and the structure of any final deal on Greek debt, have implications for the arcane workings of the CDS market. While the net outstanding volume of Greek sovereign CDS is modest at around US$3.7 billion, the current imbroglio raises important questions about the role and efficacy of CDS contracts generally.

Details of the negotiations published by Bloomberg highlighted the voluntary nature of the arrangements. Europe’s leaders summoned bankers at midnight to Luxembourg Prime Minister Jean-Claude Juncker’s office, presenting Charles Dallara, managing director of the IIF, with an ultimatum: take a package involving a 50 percent writedown of Greek debt or face worse consequences, namely the total insolvency of Greece. Bankers had little choice under the circumstances, though whether the EU would really have been willing to allow the total insolvency of Greece, resulting in the failure of some banks and large claims on the public purse to recapitalise them, will be long debated.

Similar to credit insurance, in a CDS, the buyer of protection pays a fee to obtain indemnification against the risk of default of a borrower (Greece) and any resultant loss from a protection seller. Payment is triggered by a “credit event”, technically defined as a failure to pay interest or principal, a debt moratorium or repudiation, or “restructuring”.

“Restructuring” involves a material restructuring of an entity’s payment obligations or a forced substitution of new obligations. Generally restructuring will entail any of the following:

1. Changes in the ranking of the debt, reducing seniority, subordinating the obligation or converting debt into equity;
2. A change in the currency of payment (other than into certain permitted currencies—for example, currencies of G-7 countries or OECD members with a local currency long-term debt rating of either AA or higher);
3. Any reduction in the interest or principal payable; and/or
4. Deferral or postponement in the date of payment of any interest or principal.

Restructuring is not considered to have occurred where it is not directly or indirectly related to deterioration in the creditworthiness or financial condition of the entity.

Voluntary restructuring—entailing lenders agreeing to an exchange by Greece of existing bonds and loans for ones with different terms (longer maturity, different rates)—may not constitute a credit event under the CDS. This is because lenders would be agreeing “voluntarily” to subscribe to new debt which would be used to pay off existing or maturing debt. The original debt is not “restructured” in legal terms. This means that, for Greece, only a “hard” default—a failure to pay, full debt rescheduling or non-voluntary or forced exchange—would allow the CDS protection buyer to trigger the contract.

The final arbiter of whether the Greek CDS has been triggered will be the Determinations Committee (DC) of the industry association and rule-setter for the derivatives market, the International Swaps and Derivatives Association (ISDA). The DC comprises 10 bankers and five investors. Unless backed by a supermajority of 12 out of 15 members, its decisions are externally reviewed by a committee of “independent experts”. While perfectly legal, the ability of a private body of financiers and lawyers to determine whether or not there has been “default” is unusual and legally untested.

In contrast, the rating agencies have indicated that any such voluntary exchange will be regarded as “selective or restrictive default”.

Where a CDS contract was purchased to hedge existing holdings of bonds or loans, an inability to trigger the contract will mean that investors will not be compensated for any losses on such holdings. Depending upon the terms of the exchange, the new bonds may trade at prices below par (based on the experience of previous, similar exchanges) reflecting differences between the return demanded by markets relative to the new bond’s economic terms. This will result in investors incurring losses.

All participants who purchased protection will also have incurred the cost of hedging for the ineffective CDS contracts.

So, for banks and investors who entered into CDS contracts to insure against losses from a Greek default, the potential failure calls into question the entire economic effectiveness of credit derivatives. The technical nature of the arrangements also highlights the potential legal issues present in CDS contracts. Different legal forms of economically similar actions can lead to entirely different outcomes under the CDS contract, complicating significantly the effects of the contract and its efficacy as a hedge.

As regulators and accountants assumed that the CDS eliminated or minimised any risk of losses, the level of capital and reserves set against the risks of Greek investment may turn out to be incorrect, while the accuracy of financial statements is also in doubt.

The question now is whether similar arrangements will be used if any other sovereign entity finds itself, like Greece, in serious financial distress. In effect, the value of CDS contracts is questionable. This means that risk models and hedges will need to be amended.

Following the announcement of Greece’s voluntary restructuring, many traders, both hedgers and speculators, liquidated CDS positions. In part, this reflected traders swapping CDS for short bond positions, which would have gained on the voluntary writedowns. The five year Greek CDS spread fell by more than 20 percent to close at 34.36 percent.

Ultimately, these problems raise a vital fundamental question: are CDS and, more broadly, derivatives, useful and legitimate instruments of risk transfer?

Satyajit Das is the author of Extreme Money: The Masters of the Universe and the Cult of Risk (2011). Ten copies of his book are available for free to IndexUniverse.eu readers. Please write to eufeedback@indexuniverse.com if you would like to receive one.

A version of this article previously appeared in the Financial Times.