Heykel Jelassi, chief architect back office, Murex outlines some of the issues in his SIBOS open theatre on evolving challenges in assets valuation.
SIBOS is an opportunity to address a number of factors, including the growing importance of cost of risk and funding in the investment decision making and related requirements in fair value accounting; the need to have valuation shared and properly calibrated across the different functions in an organisation to highlight divergence between accounting standards and regulatory standards; and the way decision making is impacted by multiple downstream including those related to accounting.
The most important issue is the evolution of fair value, specifically the requirement to add a number of risk and cost factors to the calculation of fair value. The key elements in this calculation are the credit value adjustment (CVA) and funding valuation adjustment (FVA), which are impacting trading decisions made by the organisation.
Fair value accounting methods were strongly criticised following the global financial crisis, primarily for the way integration of credit risk through CVA and FVA was dealt with. Previously there was a mark-to-market on trades. Now, depending on the organisation of the mark it can impact on what the traders are seeing.
Front office pricing was previously based on fair value – now it is impacted by factors such as counterparty risk. While this was already taken into account in the accounting standards (whether IFRS or US GAAP), other factors are only partially accounted for or not taken into account at all.
For the same trades of the same derivatives, different values can be used. Before the accounting, fair value was the reference from a regulatory point of view and used to already include the CVA or the impact of the credit risk of the counterparty.
The norm, especially FRS 9 (which sets out the definitions and accounting treatments for associates and joint ventures, two types of interests that a reporting entity may have in other entities) was that when you think about the fair value for a derivative you need to include the credit risk.
However, after the crisis Basel III was more specific about how to process this credit risk and how it should include the fair value, which means that we now have a difference in approach between the accounting laws, the international accounting rules and what Basel III is saying.
The accounting norms are still recommending some best practices that allow the credit risk to be offset by the institution credit risk, which in Basel III needs to be done completely separately. On the trading side, depending on whether the institution puts in place any calculations to include in its pre-pricing of this credit risk, most of the trades are still based on the fair value as it has been seen before, meaning a mark-to-market. However, mark-to-market doesn’t mean anything anymore from a regulatory point of view because you need to include this credit risk.
There are differences in the way accounting and regulation are treating this subject, which produces different valuations. For example, there is nothing in the accounting standards about how to account for funding cost, which is taken into account in the regulations. Basel II contained some elements of this but Basel III goes further in terms of what should be included in the fair value calculation.
Cost is calculated on the fair value level rather than the trade level, which means that the organisation has to decide how to calculate counterparty risk so the trader can see it. This highlights the requirement for centralisation of the calculation – if you don’t centralise the calculation (which is the case for most banks today) you end up with the accounting department and the regulation team both using their own figures.
There is no holistic approach to risk identification and calculation, which means the trader cannot be given clear indicators as to how they should include risk in their decision making. Having a global view would help us to decide whether to account for the funding cost, for example, which can be beneficial from a taxation perspective.
At the end we find ourselves with this value of credit risk that needs to be calculated differently in accounting, trading and for regulatory reporting. So it is very difficult for banks to see what are the differences on their reporting when the trader is thinking that his P&L is positive, the accountant can be showing a P&L that is negative and the regulatory reporting is using a different P&L.
I don’t think that the difference is a problem – it is between regulatory and accounting. The problem is more not being able to understand what the difference is. If you can have, for instance, figures on the accounting that are advantageous, if your P&L is not important in accounting you are basically paying less tax.
Not being able to explain what the differences are is mainly due to the technology that is used, because most of the time you have risk management that are calculating a CVA for regulatory reporting, the accounting team that are using best practices and the front office that is not at all connected to that.
On the front office side it is also very important to include all these calculations in terms of decision making – before they take their decisions they should base them on mark-to-market and how they price their trades. They need to have this calculation even before trading, so in pre-trade there will need to be indications about what are the calculations that will be reported as regulatory reporting and what would be accounted for, because the decisions are impacted by that.
If you think about just the fair value as before being a mark-to-market of a trade, if you have to pay for the amount of CVAs of course you are paying capital charges on that so the internal cost within the institution may get beyond the point at which the P&L of the trade was not that important.
In this case the decision should perhaps have been that it would have been better to trade with a different counterparty with less risk (even though the P&L that you would be having would maybe be lower) because at the end you are paying less capital net charges and the overall cost is lower.
So the technology today within institutions has to be there to enable better decision making by doing these types of complex calculations upfront. I am not referring to the complexity of the CVA but rather that calculating these indicators is a complex process and that when we were doing them post trade we had more time. Now we have more complex calculations to do upfront in order to make a decision, so we need more performance in terms of technology.
The same calculation had been used in the front office to do the most of the post trade, both for the regulatory side and the accounting side, even though they are done differently with some specificities. Being able to explain the differences enables us to have a complete view of how fair value within the organisation is seen from a different perspective.
The accounting result is that we paying the correct taxes and the trading is our financial way of looking to the P&L and an explanation is to be available for all this, meaning that the same technology should be used. If we have banks that are still on different systems to calculate and using different methods they would never be able to explain what is going on.
At the end, the decision making is completely independent from regulatory reporting and the accounting would be having completely different figures. When you add some other accounting complexities to that, such as hedge accounting (whose first objective is more towards smoothing the volatility of the P&L account) you may be taking the decision on the accounting side, but that is not completely in line with the strategy of the likely in-between points from front offices.
The impact of the CVA is of course a recommendation from Basel III, because we are saying that you have risks with the counterparties so if you have just one central counterparty, clearly the risk is less. However, the interpretation of Basel III in Europe and in the US is different. There is even more additional complexity for a multi-entity institution when they do these calculations since there are some assumptions on how the calculation should be done.
These calculations are already very complex, but now it is not just any more a matter of complexity that is the problem, but a matter first of having the form of technology in terms of calculation for the pre-trade and also integrated solutions within the bank, otherwise no calculations are possible.