Clearing, Levees, and Systemic Risk
May 13, 2015
Consider the levee – that barrier we often build next to flood-prone rivers. During a heavy rain, the levee prevents water from flooding into a nearby town, from testing and probably damaging the structural integrity of its buildings. After all, the levee has been built with only one purpose in mind, dealing with water, and therefore it does so much more effectively than most buildings.
By Modibo Camara, SAS’16
But let’s assume there is a crisis. A big storm hits, causing major flooding, enough to breach the levees (like e.g. in New Orleans after Katrina). Because a barrier was present, water was allowed to build up until it finally released in an explosive rush towards the town, magnifying the storm’s impact beyond what the town might have experienced without the levee. The levee is effective until it isn’t, and when it isn’t, it ceases to be the solution to the problem, instead becoming the propagator.
This summer, I interned at the Commodity Futures Trading Commission (CFTC), an independent government agency tasked with regulating the financial derivatives markets. At this point, you might be wondering what levees have to do with derivatives, and you’d be right to ask, because they don’t have much in common. But the levee analogy is how I’ve come to view an important new provision in the Dodd-Frank Act of 2010: the central clearing requirement. The new mandate applies a century-old practice, already common in futures and options exchanges, to various classes of over-the-counter (OTC) derivatives.
What is clearing? It’s is a process by which two parties to a financial trade (i.e. counterparties) use an intermediary (i.e. clearing house) to guarantee the contract’s fulfillment. Here’s a very stylized diagram to explain it:
The Cearing Proces
Like levees, clearing houses are equipped to deal with counterparty risk (or flooding) and allow authorities to deal with crises at a few isolated points. By pooling risk (like an insurance company) and establishing necessary safeguards against worst-case scenarios, they are more robust than individual traders, and thereby reduce counterparty risk across the system. Further, by routing all trades through clearing houses, the market is simplified from a tangled web of bilateral transactions to a multi-radial structure, wherein all trades connect directly to a few central institutions.
In response to the 2008 financial crisis, Dodd-Frank and subsequent CFTC rules now require a much larger portion of the derivatives trades to be cleared by legally designated clearing houses. Where, then, do the problems arise? In a word (or two): systemic risk. Centralized clearing does not end systemic risk, nor even counterparty risk. Instead, like the levee, it rearranges that risk. It concentrates all the pressure of a crisis onto a few key points, relying on them to protect ordinary traders – with potentially devastating results should the clearing house fail. Of course, the typical clearing house has (and is required to have) substantial risk management provisions, including default guarantee funds, high membership standards (all cleared trades have to go through approved institutions), collateral/margin requirements (more on that in a bit), and reliable short-term creditors. That does not make them immune, however, and a clearing house can be put under extraordinary stress during a market calamity. After a major price movement, counterparty failures are likely to be correlated and occur on one position of a trade, which would leave the clearing house with a large open position at seriously disadvantageous prices. Moreover, even if the clearing house can withstand an initial movement, it might be vulnerable to subsequent shocks after having depleted its emergency funds.
Given the economies of scale inherent in clearing, each individual clearing house is likely to be large and systemically important – several of them have already been designated as such by the Financial Stability Oversight Council (FSOC). The failure of a clearing house would affect all parties – not just those who traded with defaulting members, guaranteeing the spread of contagion across the industry. The shock of its failure could panic traders, and a cessation of its operations could (literally) freeze the marketplace and all assets therein. This is precisely the sort of issue that regulators have been trying to prevent in the wake of 2008. Moreover, it is not just hypothetical. Clearing house failures have occurred several times in recent history, the most famous example being the Hong Kong Futures Exchange collapse in 1987.
However, even a solvent clearing house could cause trouble during a crisis. Clearing houses generally collect liquid collateral (e.g. cash) from counterparties, to be offset losses in case of their default. Many reserve the right to demand extra collateral when the market is especially volatile. Indeed, it appears sensible for a clearing house to demand an extra buffer against losses when market-related risks are higher. However, the effect of all clearing houses demanding extra payments can put great stress on already-pressured financial markets, extracting liquidity precisely at the time when it is most needed.
There are a few other issues related to clearing, but they seem to be less problematic. One is moral hazard – just like the presence of levees could encourage lax building standards, traders might be less careful in choosing counterparties if they expect the clearing house to protect them from default. Another is competition between clearing houses, and a possible vicious cycle of deteriorating risk management practices in order to capture more market share. However, the CFTC has taken steps (particularly, in Part 39 of Title 17 of the CFR) to mitigate this possibility and to monitor excess risk-taking.
Overall, I don’t believe that Dodd-Frank’s clearing requirement was bad policy – nor do I have anything against levees. Aside from clearing houses being better equipped for a crisis, there are real advantages associated with this new distribution of risk. It’s far easier to assess market distress and provide liquidity support to a few ailing clearing houses than to a highly interconnected network of traders. Furthermore, some have compared the clearing house can act as a sort of disinterested market watcher – it has a stake in the stability of the system, but, ideally, cannot profit off distorting prices or manipulating markets. And as mentioned, FSOC has already designated several clearing houses as ‘systemically important’, and the CFTC and other regulators have taken important steps to ensure active oversight and appropriate standards among clearing houses.
If well maintained and regulated, with appropriate emergency support when necessary, clearing houses will simply be a routine part of our financial plumbing. The biggest threat today, as in the mid 2000s, is complacency: clearing does not ‘fix’ systemic risk, and it must be watched, but under due vigilance it can do a lot to improve our financial system.
Disclaimer: these thoughts are entirely those of the author, were written privately, and do not represent or reflect the views of the CFTC or its staff.
 To get an idea of the size of these markets and their clearing rates today, see the CFTC’s weekly swaps report (http://www.cftc.gov/MarketReports/SwapsReports/L1GrossExpCS). For those new to the topic, however, please note that notional value almost always overstates the value that actually changes hands.
 See the 2011 speech delivered by Paul Tucker at the Bank of England, on page 4 (http://www.bankofengland.co.uk/publications/Documents/speeches/2011/speech501.pdf). Also see Ben Bernanke’s 1990 paper, Clearing and Settlement During the Crash (http://www.jscc.co.jp/en/data/5), which discusses stresses faced by US clearing houses during the same crisis and emergency actions taken by the Fed.
 See Craig Pirrong’s 2011 paper, The Economics of Central Clearing: Theory and Practice. It was written under the ISDA, and covers many of the topics that I mention here, only at far greater depth.
4 This also raises the question of why central clearing can’t be a public service, where the clearing house is an explicitly guaranteed federal institution. At the least, this would avoid any principle agent problems related to clearing houses’ dual role as for-profit corporations and system stabilizers. Interestingly, a 2014 speech delivered by Paul Tucker, entitled Are Clearing Houses the New Central Banks? considers the idea (http://www.chicagofed.org/digital_assets/others/events/2014/annual_over_the_counter_derivatives_symposium/tucker_clearinghouses_new_central_banks_tucker_2014.pdf).
 It has been pointed out to me, however, that since clearing houses thrive on volume, they may have an incentive to distort the market in that direction. As with any institution, particularly for-profit ones, their interests will never fully align with those of society.
Student Blog Disclaimer
The views expressed on the Student Blog are the author’s opinions and don’t necessarily represent the Penn Wharton Public Policy Initiative’s strategies, recommendations, or opinions.