Skip to main Content
Site Search

Advanced Search

  • Mondo Visione
  • Mondo Visione - Worldwide Exchange Intelligence
Member Login

Member Login

Forgotten your password?

Managing post-trade risk

Date 29/10/2008

Tim Reucroft, Thomas Murray Ltd

So you’ve read the book and decided which exchange and product to trade – but what about the post-trade environment? This is where the complications start: account openings, funding clearing and settlement accounts, delivering securities or commodities, optimising cash flow through the various payment mechanisms and ensuring any collateral is in place.

Does the agent you’ve appointed handle all the markets you want to trade and do they do it directly themselves or lay off the risk to third parties? What if you want to do cash products, exchange traded and OTC derivatives – do you need separate arrangements for each of these? How does your agent handle cross-product and cross-border post-trade arrangements?

Too many questions to answer, but we can approach the subject generically – what are the risks and where are they going – up or down? The main categories of risk that we need to consider are:

  • Market risk
  • Credit risk
  • Operational risk
  • Legal risk.

Post-trade risks

Once a decision to trade has been executed, the market risk is assumed. The post-trade risks (which include an element of credit risk, plus operational and legal risk) cannot then be avoided, only managed. There are factors working both ways:

  • Increasing risk: product complexity, product diversity, increasing volumes and emerging markets.
  • Decreasing risk: functional convergence, consolidation of utilities and the retail settlement model.

Without a measure of these risks it is impossible to quantify if the overall risk is increasing or decreasing. While we have a very good measure of market risk and a means of mitigating credit risk, we are still in the dark as regards operational risk and legal risk, although the regulators are constantly pushing for the market to measure and manage these risks. It started with market risk back in the 1990s when various models were tried until Value-at-Risk (VAR) supported by risk metrics was universally adopted. Next came credit risk when the regulator adopted the same tactics of pushing the problem onto the market. While credit risk might be more difficult to quantify (the sub-prime crisis springs to mind), it became easy to manage with the advent of credit default swaps.

These were so successful that they created an unacceptable level of operational risk which the regulators pushed back on to the market to sort out. (This solution is extremely interesting in the context of functional convergence). The regulator now wants the market to measure and manage operational risk (or at least set aside capital) so it is only a question of time before somebody comes up with a mechanism to transfer operational risk between counterparties. (A bit like carbon permits, only giving permission to make mistakes). Will the introduction of operational risk measures create operational risk, as credit default swaps did? probably. How long before the regulator requires the market to measure and set aside capital for legal risk? An interesting question that may be influenced by Target 2 for Securities (T2S).

To manage risk you create the equal and opposite. This means some form of derivative – but this itself creates risk.

Increasing risk

Product complexity

There is no doubt that products are becoming more complex. The driver for this lies with the institutional investors who are now pursuing alpha rather than beta. The proliferation of indices, exchange traded funds (ETFs) designed around these and warrants all provide much wider investment choices but add to the complexity. However, it is the explosion in OTC derivatives that has exponentially leveraged the risk. Bespoke products across a range of asset classes and swaps across asset classes have created a post-trade nightmare. Add to this credit default swaps and it was little wonder that back offices could not cope.

Product diversity

The search for alpha has driven investors into alternative asset classes as the supply of alpha in the cash markets became exhausted. Seismic shifts in the global economy, driven by raw material shortages, forcing the switch to substitute goods (corn, palm oil, sugar), with an overlay of global warming threatening current supplies, has generated a surge in the agricultural economies. This has created its own alpha and enticed institutional investors into commodities. Derivatives are on hand to meet the demand for carbon permits, weather futures and uranium supplies. Property and private equity are also on the shopping list to fill out the product diversity.

Increasing volumes

Institutional investors were not the only ones seeking excess returns. The investment banks discovered algorithmic trading (computer-driven trading that forced the Chicago exchanges to impose position limits on derivatives traders back in the 1980s). Estimates vary but up to 60% of trading volumes can be attributed to algorithmic traders. Even so, in the volume stakes the pecking order is exchange traded derivatives, stock lending and borrowing, with the equity markets a poor third. Stock lending and borrowing creates its own post-trade issues with settlement in the form of Delivery by Value (DBV) rather than Delivery versus Payment (DVP).

If all the transactions emanated from a single source (global exchange consolidation) then the operational risk would decrease (but at the expense of concentration risk). However, what we see at the moment is the opposite; consolidation of traditional exchanges but only at the legal entity level, not operationally, plus a plethora of Multilateral Trading Facilities (MTFs), often with their own discrete post-trade operations. Bringing on stream these new infrastructures will inevitably increase the operational risk. If the new MTFs succeed in driving down the execution costs then this will expose the settlement costs of a transaction. This will encourage consolidation in the post-trade space. An example would be T2S where consolidation (might) reduce the costs but at significant operational risk (during implementation) and at the expense of enormous concentration risk.

Emerging markets

Today’s emerging market was yesterday’s frontier market. In the drive for enhanced returns, investors are continuously opening up new markets. A few years ago Brazil and Mexico might have been the only South American markets that investors felt comfortable with, but today you would add Chile, Colombia and Argentina, with only Venezuela regarded as uninvestable.

Decreasing risk

Functional convergence

Imagine trying to settle cash equities on a bilateral basis without a CSD. The operational risk would be enormous. Imagine if exchange traded derivatives were not novated to a CCP and therefore were not fungible and had to be settled bilaterally – the volumes would be enormous. Netting by novation and settling net cash movements via a CCP significantly reduces operational risk.

However, in the OTC derivatives world transactions were being settled on a bilateral basis until DTCC stepped in with Deriv/SERV, a central matching and settlement system (via CLS) for the credit default swaps market. Later a trade information warehouse was added to store details of all the transactions so that DTCC could also manage the revaluations, resets and margin movements. By late 2007 cash netting became available via CLS, significantly reducing the value and number of payments.

What we are seeing is an example of convergent evolution – similar morphology derived from very different phyla. The cash markets now pass their transactions through a CCP to gain the benefit of netting (the reduction in counterparty risk for a settlement cycle of T+3 is not worth the effort). The functional solutions developed over many years in the mature cash and ETC markets are now being applied to the OTC market. The key is an infrastructure solution – a utility that the market shares to reduce the risk – confining the operational entropy. Candidates are existing exchanges (CME Swapstream), CSDs (DTCC, Euroclear), CCPs (LCH’s Bclear) or newly formed ventures (Swapswire, TZero, Tri Optima).

Utility consolidation

At the present time the developments that have taken place in the mature markets are being repeated in the new(er) OTC derivatives market. The issues of ownership, for profit motives and competition vs consolidation amongst the newly formed services is merely repeating the issues that the CSD world went through 5 years ago. The first move in this game was the acquisition of Swapswire by Markit. For CSDs the outcome is clear – consolidation, not competition.

We are seeing the same theme in the exchange world but the CCPs are busy in a price war, weakening the competition in preparation for consolidation. The differences are that CSDs are natural monopolies, CCPs are not, and the exchanges have traditionally had a monopoly until MiFID. Thus is makes sense for the CSDs to consolidate, since competition is quite awkward. It makes sense for the exchanges to consolidate to pool liquidity (even more so now that they need to compete with the MTFs), but the bitter battle in the CCP world looks likely to run its course (think of the risk consequences of a ‘race to the bottom’).

It will be interesting to see if more utility solutions emerge in a competitive environment, or if the major providers already in this space will seize the opportunities and move straight to a global monopoly. What is clear is that exchanges should take a very keen interest in this space. The debate over the role of traditional exchanges is being pushed hard by the new MTFs and the cash market exchanges are looking downstream to secure their future. Acquisition of the CSD is an obvious strategy and although vertical silos were out of favour for a while, as the EU Commission branded them anti-competitive, they are now back on the agenda – Switzerland and Norway both opted for vertical silos in 2007.

A problem in many markets is identifying who is responsible for utility consolidation. In emerging markets there is pressure to release the utilities from governmental control (where this is often the case), but this creates a vacuum as to who might be responsible for taking the disparate components forward. Often the regulator will take a lead, but they might be viewed as quasi governmental (CSRC in China, although SEBI in India is an excellent example of a regulator moving the market forward), while another approach is for the market to sign up to a capital market’s modernisation plan (examples including Malaysia and Hong Kong). The UK has been a disaster from a utility perspective – most of the components have been sold (CREST, LIFFE, LCH), while the Germans and French have done extremely well. This perhaps reflects the ambitions and vision of the individuals running these infrastructures, something which was lacking in the UK – no one individual ever had the responsibility.

The retail settlement model

The major market dichotomy is between those that operate the retail settlement model and those that operate the wholesale settlement model. Essentially, a CSD operating the wholesale model might maintain several hundred omnibus accounts, whereas a CSD operating the retail model would typically have several million accounts, all at the beneficial owner level. The wholesale model was created in the 1970s to reduce the operational risk arising from the paper mountain in the USA when settlement volumes swamped the market. At the time omnibus accounts were a good solution to the problem; technology was expensive and a single omnibus account reduced the reconciliation effort and the number of instructions.

What they did not consider was the legal risk associated with omnibus accounts. Without understanding the legal detail it is sufficient to realise that UNIDROIT, the intergovernmental organisation based in Rome, funded by 61 member states, has been charged with resolving the legal issues associated with the wholesale model (termed indirectly held securities by UNIDROIT) since 2002, but has been unable to come up with a solution. Clearly, the issues are not trivial. An interim solution came via the Bernasconi Report in July 2006 which recommended adoption under the Hague Convention.

What is apparent is that the majority of the legal issues with the wholesale model do not arise in the retail model. Many new markets have taken the opportunity to install the retail model as technology is significantly cheaper than in the 1970s. Volume is not an impediment; SD&C, the CSD in China operates approximately 120 million accounts where securities settle on T+0. To do this requires capturing the beneficial owner at the point of trade – hence the exchanges interest in this model. Where the retail model operates there is clearly a role for exchanges as part of the information flow and synergies where the exchange and CSD are under common ownership.

It is staggering that the ECB is pushing for all Euro zone markets to adopt the wholesale model since T2S cannot cope with the retail model. As Shakespeare’s Hamlet said,“There is something rotten in the state of Denmark” (actually Denmark is OK because they have the retail model). Thus the exchange’s involvement in the retail information flow brings them closer to the CSD and makes a vertical silo a rational objective. While such an arrangement might be used to protect the exchanges franchise from MTF competition (since they control the downstream settlement), the response by the MTFs is to create their own settlement arrangements, thereby compounding the risks.

The Thomas Murray approach to post-trade risk

Several years ago, Thomas Murray developed a model to monitor post-trade risks in conjunction with a number of custodians, CSDs, the World Bank and Standard & Poors’. We have been applying this model ever since, acting as a global clearing house for information affecting over 100 markets and over 140 CSDs. We maintain securities market profiles on over 100 markets detailing the local rules on trading, clearing, settlement and custody plus supply risk ratings, both private and public, for the CSDs and for 88 markets.

The post-trade risks we assess to make this judgment consist of:

  • Asset commitment risk
  • Liquidity risk
  • Counterparty risk
  • Asset servicing risk
  • Financial risk
  • CSD on CSD risk
  • Operational risk.

For a description of the risk categories and examples, see our web site: www.thomasmurray.com/csd-ratings/central-securities-depository-ratings.html.

Given that CSDs are essentially designed as risk minimising vehicles there are still a considerable number that leave room for improvement. We still find that a number of our market flashes carry a negative risk assessment. For the year 2007 the distribution was as follows:

Negative assessment 15%
Assessment on watch 35%
Positive assessment 37%
Stable assessment 13%

Thomas Murray Depository Service 2007 newsflashes by outlook

 

Negative

On watch

Positive

Stable

Total

Asset commitment risk

12

14

3

29

Asset servicing risk

1

2

17

6

26

Counterparty risk

7

29

63

13

112

CSD on CSD risk

9

8

4

21

Financial risk

98

92

30

33

253

Liquidity risk

5

31

100

27

163

Operational risk

15

110

67

21

213

Total

126

285

299

107

817

Source: Thomas Murray, 2008

Outlook scale

Stable: There are no factors at this time that would affect the capital market infrastructure risk rating

Positive: Factors that may result in an improvement in the capital market infrastructure risk rating

Negative: Factors that may result in a deterioration of the capital market infrastructure risk rating

On watch: Factors that may result in a change in the capital market infrastructure risk rating, but the direction of the change is uncertain at this time.

These risk assessments only cover cash market operations; they do not assess the post-trade risks associated with derivatives, although a number of CCPs that act for derivatives also act for the cash market. While CCPs that act for derivatives do so in a reasonably robust manner, the same cannot always be said for the (same) CCP acting for the cash markets. There are numerous ‘CCPs’ that claim to offer risk elimination, but do so by passing it back to the participants rather than absorbing the risk themselves, or alternatively only covering the replacement cost, not the full principal. We call these CCP Lite.

Once we know the component risks we can bundle these into a risk metric aimed at a particular product. Thomas Murray has developed a set of three Market Infrastructure Investability Indicators (MIIIs) used in the construction of the GEMX index that will be used to benchmark the World Bank's GEMLOC fund investing sovereign bonds. The three MIIIs cover over 20 emerging markets for: (i) the efficiency of clearing and settlement systems; (ii) the safety and soundness of safekeeping arrangements; and (iii) the efficiency of asset servicing. Assessing the post-trade risks becomes a little easier.

Conclusion

The financial sector makes money by assuming risk. Excess risk will eventually attract excess capital (once the regulator catches up) so that the risk adjusted returns remain comparable across the various product lines. To capture above average returns therefore often means accessing emerging markets where the risk/reward profiles are out of line with the developed countries. While the rewards in emerging markets are quite obvious, the risks are often hidden and require significant due diligence to uncover. A well understood model taken from the developed markets does not always translate exactly to an apparent similar model in the emerging markets and it is the subtle difference where the risks reside. Harmonisation via the adoption of global best market practices and international standards is one way forward but the optimal solution is consolidation. The consolidation that is currently coalescing markets will force through the required harmonisation and make the management of the post-trade risk less hazardous – one less dimension to worry about.