- RSS Channel Showcase 7451961
- RSS Channel Showcase 9222347
- RSS Channel Showcase 2148100
- RSS Channel Showcase 8406191
- 04/09/15--05:26: REVISION: BLAS Extensions for Algebraic Pricing Methods
- 04/26/15--22:26: New: Capital and Funding
- 10/01/15--22:56: New: The Cost of Clearing
Banks hold and routinely exercise the option of freely re-hypothecating variation margin across counterparties and trades. However, the emerging FCA/FBA standards for funding cost accounting are mostly formulated in terms of netting set specific metrics that fail to properly account for re-hypothecation benefits to Common Equity Tier 1 Equity Capital (CET1). Additionally, the FCA/FBA standard introduces a double-counting issue between funding benefits and DVA which ultimately leads to a violation of the fundamental accounting tenet of asset-liability symmetry.
In this paper, we propose an alternative accounting framework meant to rectify some of the problems in existing standards. The new accounting method, denoted FVA/FDA, explicitly incorporates the re-hypothecation option into its definition of funding costs, and maintains consistency with the Modigliani-Miller Theorem, with fair-value asset-liability symmetry and with Basel III rules for DVA and equity capital. We argue that ...
We recently introduced the FVA/FDA accounting framework for funding costs, aiming to provide an accounting method that reasonably balances the, often conflicting, concerns of accountants, regulators, traders, and financial economists. While introduction of FVA/FDA accounting does not lead to write-offs of Net Income, regulatory capital measures and FTP policies are considerably different from those in funding-free accounting regimes. In this paper, we provide a concise comparison of FVA/FDA accounting with the FCA/FBA method currently endorsed by several large banks. We discuss detailed FTP policies, risk management implications, and quantify the notion of funding arbitrage.
PDE pricing methods such as backward and forward induction are typically implemented as unconditionally marginally stable algorithms in double precision for individual transactions. In this paper, we reconsider this strategy and argue that optimal GPU implementations should be based on a quite different strategy involving higher level BLAS routines. We argue that it is advantageous to use conditionally strongly stable algorithms in single precision and to price concurrently sub-portfolios of similar transactions. To support these operator algebraic methods, we propose some BLAS extensions. CUDA implementations of our extensions turn out to be significantly faster than implementations based on standard cuBLAS. The key to the performance gain of our implementation is in the efficient utilization of the memory system of the new GPU architecture.
This paper sets out a framework to manage and account for the costs of capital and funding through a pair of KVA/FVA metrics. While the FVA is the cost of unsecured funding, the KVA depends on a hurdle rate chosen by management to compensate shareholders for earnings' volatility and the risk of dilution due to recapitalisation in the event losses are absorbed by capital.
The definition of the KVA depends on an Economic Capital model to quantify capital at risk. To avoid day-one profits, the KVA is provisioned at trade inception and released gradually into earnings on a mark-to-market basis. Recognizing capital as a source of funding, we re-define the FVA to avoid an overlap with the KVA.
Clearing members face substantial costs for capital as they post Initial Margin (IM) and are required to contribute to the Default Fund (DF) of the Central Counterparty Clearing House (CCP).
In this paper, we discuss how these costs can be priced and passed on to clients. A transfer pricing policy entails three elements: the cost of capital for the Default Fund Contribution (DFC), the MVA or present value of IM posting obligations and the cost of counterparty credit risk against the exchange.
The DFC is particularly delicate to cost because clearing members cannot predict the size of their contribution as this depends on the global CCP portfolio. We derive a pricing benchmark by assuming that CCPs collect incremental KVA at each trade and remunerate DFC at a hurdle rate matching that of clearing members.