P&L Attribution analysis and reporting provides users with a coherent breakdown of the drivers of P&L movements between two points in time with reference to a select number of easily understandable pricing factors. P&L Attribution can be calculated in two ways, either the Sensitivities method or the Scenario based method.
The Sensitivities Method involves the calculation of a trade’s sensitivities (also known as the Greeks) and then using them to predict the expected change in P&L from one period to the next by using the actual market changes in the factors driving the transaction price over the same period and the transaction’s sensitivity to those factors. For example, let’s assume the value, V, of a trade is affected by three variables, time τ; underlying price p and another generic factor, x:
If we call the trade value V’s sensitivity to a change in the underlying asset price ‘Delta’ or ‘δ’ and the observed underlying asset price p at time t is ‘pt‘ then the expected P&L change due only to the change in underlying asset price can be approximated as follows:
Impact of underlying price on value V ≈ δ.(pt-1 – pt-2)
To calculate the total expected impact on the trade value we simply repeat this approach for the other factors which drive the value of the transaction and then aggregate these to give a final expected value change. This predicted value change can then compared to the observed change in the transaction’s value.
The Scenario or Revaluation method executes a series of valuation scenarios, one for each factor driving the transaction value, whereby each scenario changes one factors’ value at a time by an amount equal to that observed between the two time points in question. P&L attribution is then calculated by aggregating the impact of these valuation scenarios and not on fixed sensitivities. Therefore,
Impact of the change in Time = V(τt – 1,pt – 2,xt – 2) – V(τt – 2,pt – 2,xt – 2)
Impact of the change in Underlying Prices = V(τt – 2,pt – 1,xt – 2) – V(τt – 2,pt – 2,xt – 2)1
The Scenario method will tend to be more onerous on computing time and resource than the Sensitivities Method but it has the advantage of more accurately taking into account second order effects, such as gamma.
For each method the expected P&L may be different from the actual P&L, this will then give a ratio of explained to unexplained P&L.
1 strictly speaking, pt – 2, should actually be the forward value of pt – 2 at time t – 1.
To find out more about P&L attribution: contact Peter Griffiths on our P&L attribution teamBack to P&L Attribution main page
Why do it?
P&L attribution offers a important control, the decomposition and analysis of actual booked P&L and its variance to that of a risk based theoretical P&L provides a daily test of the models. In addition to a risk control it also provides an valuable operational control of trade amendments and cancellations by highlighting their effects within the explain.
Regulators are increasingly focused on enhancing the control environment within Investment Banks and Asset Management firms. The linkage of risk sensitivities (“Greeks”) to the real world profit and loss figures provides an additional control framework. In the May 2012 Consultative Document from the Basel Committee entitled “Fundamental Review of the Trading Book”, the committee proposes that in its revised assessment of portfolios eligible for treatment under the internal model approach for risk weighted assets that:
“Where a trading desk does not achieve acceptable P&L attribution or back testing results, the bank would be required to calculate capital requirements for that desk using the standardised approach.”
Given the increasing demands on Bank capital, a solid P&L attribution framework will continue to be a key driver of organisational capital efficiency.
The Dangers of Unexplained P&L
Large unexplained P&L balances, i.e. where sensitivities multiplied by market movement fail to reconcile closely with reported P&L, should provide internal control functions with warning signs as to the validity of risk and P&L methodologies. This can be a further challenge where traded instruments are illiquid and little or no market observable pricing exists.
P&L Attribution Dynamic Reporting
The proper display and visualisation of the P&L Attribution data is a vital step in the process to ensure timely and accurate interpretation of the results and risk. Calimere Point Risk Advisory is an expert in all forms of risk information reporting. We provide our clients with sophisticated reports allowing them unparalleled insight into the high level and low level risks they are running. Our objective is to provide users with information, fast. Typical implementation times are measured in weeks and months, not years.
Our optimised P&L Attribution reporting allows users to quickly identify the primary factors causing movements in P&L through a simple and easy to understand interface incorporating both numerical and graphical data representations
Draw visual comparisons fast
Reports incorporate graphical visualisations of each identified attribute at report date and also historically through time with changes.
Large quantities of historical information at your fingertips
Our streamlined report interface allows for the historical performance of all P&L Attribution factors to be tracked through time and the ongoing monitoring of current and historical unattributed P&L and unexplained risk factors.
Designed By Risk Professionals For Risk Professionals
We offer clients a pre-defined but customizable P&L Attribution reporting interface that incorporates numerical and graphical representations and offers a consolidated view of all key risk factors relevant to the products under consideration
Get The Information You Want – Now
Users can define the combination of P&L Attribution factors they wish to interrogate depending on the purpose of the analysis, and sort values to identify largest relative or absolute exposures using just a couple of mouse clicks
Pinpoint Problems By Drilling Down From High To Low Level Information
Our solution allows P&L Attribution to be viewed at all levels of aggregation, from country or enterprise level down to division and desk and ultimately each individual trade, moving between levels with a single click
To find out more about our P&L attribution reporting applications: contact Ian Fernandes on our risk reporting team
Data Deep Dive
In organisations where risk and P&L reporting have been typically siloed, significant discipline is required in the capturing, managing, cleaning and verification of diverse sources of data. A data “Deep Dive” assisted by CP Analytics will trace the relevant data from capture to calculation to ensure data and methodologies are consistently applied.
For example, data corruption can occur between systems, numbers can be modified or trade attributes overwritten. Sensitivities may be based on a 1 basis point movement in one business area and a 1% shift in the underlying in others, or in some cases parallel shifts rather than specific curve bumps may be used to capture particular sensitivities. Failure to accurately capture these differing approaches can lead to a large unexplained P&L.
The P&L Attribution process is heavily dependent on different upstream data sources, FO risk engines, market data sources, valuation batches and finance reporting tools. In the case of an upstream failure, the attribution process needs to be able to respond intelligently to failure and only reload failed dependencies, limiting the impact on the P&L Explain clients. In many cases existing attribution processes are complicated and require the whole process to be re-run after the upstream failure is cured, affecting all attribution clients.
CP Analytic’s attribution process is represented in a visual workflow that clearly identifies the specific points of failure and affected processes. This will ensure that unaffected processes can be completed and that only affected processes requires remediation.Back to P&L Attribution main page