Guest authors: Alexander Burck, Head of Corporate Financial Controlling, Bayer AGDr. Sven Ludwig, Managing Director, Head of SME Risk Management and Analytics EMEA, FIS / SunGard
This contribution reflects the opinions of the authors. Furthermore, it should be noted that all information published by the authors reflects their own personal opinions only and should and must not be construed as reflecting the opinions of either of their employers.
Due to ever changing market standards and tighter banking regulations, hedging with OTC derivatives, i.e. derivatives traded over-the-counter, has become more complex. This is evident even to uninvolved observers from the large number of 'new' acronyms in this environment, such as CVA, DVA, FVA, KVA etc., collectively referred to as XVA. These adjustments of familiar OTC derivative measurements of various causes are brought into context in this paper for companies in the real economy. We also explain how these XVAs are interconnected.
At the same time as establishing these adjustments of various causes, issues such as executing OTC derivatives transactions via central counterparties (CCPs) need to be addressed. These changes affect non-bank financial institutions of all sizes both directly and indirectly. This development is directly noticeable in Treasury, as the OTC derivative's price usually increases on conclusion of a trade and the pricing structure of banks seems to be becoming less transparent and comparable overall.
Companies reporting in accordance with IFRS in particular have to disclose credit valuation adjustments (CVA) and debt valuation adjustments (DVA). The International Financial Reporting Standards require consideration of all risks during fair value measurement1. Only an adjustment of such measurement by including the counterparty credit risk of both parties, also referred to as bilateral CVA, would be in line with these fair value requirements. However, these valuation components are still determined in a rather unsophisticated manner on an approximate basis in accounting in many cases and not where they arise directly and noticeably: in Treasury.
This article addresses several of these challenges caused by largely unintended consequences of these changed market standards – challenges which companies in the real economy have to face now and probably increasingly so also in the future.
Background to changes in current market practice
Until the major financial market crisis in 2008, many players in the financial markets valued their derivatives at market prices without explicitly taking account of counterparty credit risk. This risk was particularly not evaluated in cases where the counterparty was a large internationally operating bank that was "too big to fail", which frequently also received a better rating than its counterparty. Banks also operated under the "too big to fail" premise amongst each other.
This attitude has only changed since the de facto or actual failure of major financial institutions such as Lehman Brothers, Bear Stearns, AIG, Fannie Mae, Freddie Mac and large monoliners. Since then, CVA has become a clear focus of practitioners, regulators and academics (CVA is the difference between the risk-free portfolio value and the true portfolio value that takes into account the possibility of a counterparty’s default2).
Nowadays, all major financial institutions appreciate the necessity and also benefits of measuring and evaluating CVA and even XVA. These institutions therefore invest money and resources to improve their methods for evaluating and measuring the various aspects of counterparty credit risk exposure. Determining XVAs by means of a complete and computationally complex Monte Carlo simulation has become the established market standard.
These calculations are based on determining a derivative's present value for as many future points in time as possible. The market prices applicable to these dates for measurement purposes are determined via future pathways simulated in Monte Carlo (typically more than 10,000). For an interest rate swap extending over ten years, typically more than 100 future points in time (1 day, 2 days, 3 days, 1 week …, 10 years) are required for adequate simulation3. For example, 10,000 scenarios result in a million (100 x 10,000) measurements4.
Legislators are attempting to stabilize financial markets. This is to reduce the impact of counterparty default on financial market stability. A core instrument of European Market Infrastructure Regulation (EMIR) is the implementation and incremental expansion of mandatory execution of OTC derivative transactions via central counterparties (CCPs) such as Eurex Clearing or LCH.Clearnet. This requirement applies to certain instruments if all trading partners for these transactions are banks.
Counterparty risk is mitigated through continuous exchange of exposures with the central counterparty throughout the day. The amount of collateral is calculated by the central counterparty as the sum total of two components, the variation margin and the initial margin. In simple terms, the variation margin corresponds to the current market value of the derivative. The initial margin, on the other hand, is the prospective component to offset potential losses in the course of liquidation, in the event of default of the counterparty or client of the CCP.
The initial margin therefore is a VaR-based calculation. Both components also change during the day, even in the case of stable portfolios, simply because of fluctuations in market price. This collateral/margining concept is designed for a nearly perfect collateral agreement in bilateral scenarios. It should be noted that the overall margin is always equal to the derivative's current present value or higher.
In addition, the legislator will increase the cost of counterparty credit risk exposures for financial institutions considerably at several levels. Capital requirements will rise significantly according to the new standardised approach for measuring counterparty credit risk exposures (SA-CCR) and will increase dramatically in some cases for uncollateralized OTC derivative contracts5. The capital adequacy requirements for risk exposures serve the purpose of reducing the probability of insolvency of banks as a result of unexpected losses to a defined minimum. When viewed in isolation, this is to the advantage of the real economy.
Putting things into a Treasury context
The greater the pressure on banks due to capital adequacy requirements, the greater their pressure to execute OTC derivatives transactions with the real economy via central counterparties. This means that the gap between set prices for centrally executed OTC transactions and bilaterally executed uncollateralized alternatives (without collateral agreements) will become ever greater, and significantly so.
The greater the difference between the credit rating of the non-financial corporation and the counterparty financial institution, the greater the CVA premium payable. As the CVA is virtually non-existent in the case of a perfect collateral agreement or centrally executed transaction, the potential savings are assumed to be greatest in such a scenario. This saving and mitigation of credit risk (default risk) is offset by the cost of providing collateral (typically cash or securities) in the required amount. The anticipated cost of generating liquidity throughout the entire life cycle of a derivative − in simplified terms − is referred to as funding valuation adjustment (FVA)6.
If the derivative's market value is positive, collateral is received, if it is negative, collateral has to be provided. This leads to company-specific funding costs for both counterparties7. These costs and further implications presented in the following paragraph need to be compared with the savings from CVA. In addition, banks have to assess the anticipated cost of regulatory capital requirements, the so-called capital valuation adjustment (KVA). This cost is passed on directly or indirectly to the counterparty.
To conclude this simplified description, one should be aware that counterparties usually have netting agreements. These lead to the consequence that the present values of derivatives can be netted with those of the counterparty (transactions within a netting arrangement or netting sets) for the purpose of XVA calculation. This makes the calculation of XVA indicators and thus a derivative's present value dependent upon the counterparty or portfolio.
This is easily illustrated by the following example. A matching counter transaction to an existing portfolio has the effect that the present value of this portfolio is nil at any time. This also eliminates the CVA and FVA, setting them to nil. Concluding an identical transaction with a new counterparty would generate a positive CVA and FVA, however, and of course not entail a change in the XVA measurements of the existing portfolio. This dependency on the portfolio, in turn, has far-reaching consequences itself, which are presented in the following paragraph.
A selection of implications for companies in the real economy
A simple price comparison of potential OTC transactions beyond potential counterparties (banks) is no longer feasible due to the dependence on portfolios. The measurement components (CVA and FVA) need to be shown separately. This is the only way to compare set prices and prove market conformity for compliance purposes upon conclusion of a transaction. Understanding CVA and FVA therefore is also important for companies in the real economy, and not only to strengthen their own negotiating position.
The above explanations neglect to address the fundamental purpose of hedging and consideration of funding sources in the non-financial environment. The purpose of hedging sources of funding is to hedge cash flows from real production, payment and delivery obligations. The matching position is usually not a derivative. If the financial hedge results in a hedging obligation, sufficient collateral (cash or securities) needs to be set aside for that purpose. The amount of collateral depends on the potential and unknown performance of the derivative.
The maximum collateral required cannot be determined unless a limit is set for the present value of the derivative, while it is very well possible to estimate the maximum collateral required for a particular confidence level. This is done by means of so-called potential future exposure (PFE)8. What is not clarified this way is the source of collateral, if the required collateral exceeds the calculated potential future exposure.
Therefore, collateral has to be set aside in not insubstantial amounts. The reason for this are precisely those future scenarios that do not require hedging in the real economy, because companies generate additional profits from the underlying transaction. The funds to be set aside are then not available for operations. Moreover, margining creates unpredictable cash flows, so that cash flow matching between the derivative and the hedged item becomes virtually impossible.
A selection of implications for treasury management systems
Firstly, it should be noted that the requirements and importance of risk management and controlling are becoming ever greater and the possibilities of treasury management systems should be analysed and reassessed in view of the change in market practice. Support for the more sophisticated calculation of XVA is a critical component to be considered, more accurate control of risk is another. Limiting counterparty credit risk with a view to the future is given additional focus.
Due to the extremely large number of necessary calculations outlined above, the computing power to be made available needs to be considered in its own right, which is only achievable with modern technologies. Besides calculation, analysis of XVA is an important step in this process. Depending on the organizational structure, the treasurer must gradually be given more access to analytical possibilities, as the associated risks of new market practices become the direct focus of front office functions.
Ultimately, data requirements will increase in both dimensions – quantity and quality – so that transparency and verifiability will have to be considered separately.
Robust calculations and automation of all process steps from business initiation to (hedge) accounting are essential, in order to be able to draw benefits from increasing complexity and to ensure that the entire finance department remains efficient.
In addition to the extensive reporting requirements and necessary infrastructure to that end, as well as resources required under EMIR and comparable regulations worldwide, hedging with OTC derivatives has also become considerably more complex. XVA indicators will help to better understand and assess this new world. The instruments used to increase financial market stability diametrically affect the stability of the real economy.
In order to also develop adequate hedging strategies in the future, it is indispensable to calculate and assess these new indicators within the familiar environment of derivatives measurement, cash flow risk and liquidity planning, to name but a few. These hedging strategies are becoming more complex, and may even require complex structured products. As counterparty structure determines both measurement through profit or loss and collateral requirements, active control of counterparty risk (including Monte Carlo-based PFE and XVAs) should also be contemplated in addition to gaining an understanding of the new measurement components.
Markets change ever more rapidly these days – making continuous investment and innovation in the area of treasury management systems indispensable. However, these new possibilities have to be competently implemented. This means that Treasury needs to be upgraded both technically and professionally.
1 See IAS 39 and IFRS 13.
2 Simply put, the CVA of a derivative financial instrument is an American-style option on exactly such an instrument adjusted by the probability of default of the counterparty and the ratio at which the option can be re-exercised. For more information, see Kontrahentenrisiko: Bewertung, Steuerung, Unterlegung nach Basel III und IFRS, Sven Ludwig, Marcus R.W. Martin, Carsten S. Wehn (editors), Schäffer-Poeschel.
3 In practice, it is important to use portfolio-specific dynamic points in time in addition to fixed points in time. This improves the accuracy and stability of calculations.
4 In addition to purely technical challenges, calibrating the Monte Carlo simulation for example is a challenging task from a professional point of view.
5 Potentially more than one hundred times compared to perfectly collateralized contracts.
6 It is immediately obvious that for determining the required collateral the first step is to calculate the derivative's present value via many future pathways, just as for calculating the CVA.
7 For the sake of completeness, it should be pointed out that central counterparties do not provide collateral, only receive it. Therefore, margining is typically unilateral.
8 The PFE can also be determined by a Monte Carlo simulation, analogously to CVA and FVA. Although, depending on the risk strategy, another (real world) calibration may be necessary.
Source: KPMG Corporate Treasury News, Edition 51, January 2016
Contact KPMG: Andrea Alilovic, Senior Manager, firstname.lastname@example.org
© 2016 KPMG AG Wirtschaftsprüfungsgesellschaft, ein Mitglied des KPMG-Netzwerks unabhängiger Mitgliedsfirmen, die KPMG International Cooperative (“KPMG International”), einer juristischen Person schweizerischen Rechts, angeschlossen sind. Alle Rechte vorbehalten.