Development of Market Risk Appetite Goal - PowerPoint Presentation

Slide1

Development of Market Risk Appetite Goal

ERCOT CWG/MCWG MeetingItem 8February 18, 2015

ERCOT

PublicSlide2

Development of Market Risk Appetite Goal

BackgroundFrom the CWG Charter, two of the assigned functions are “Providing input on credit matters as requested by the F&A Committee or by TAC or TAC subcommittees” and “Providing input regarding NPRRs that impact credit in accordance with this Charter and Board approved credit policies”, further “In carrying out its responsibilities, the CWG shall be guided by industry best practices”Credit Work Group (CWG) has set itself a 2015 goal to “clarify the market’s risk tolerance / appetite level.” A rationale for the goal is to enable a framework for reviewing the risk impact of NPRRs or other market changes relative to current market design.In its December

2014 meeting

the F&A Committee requested additional

guidance on how market risk might be assessed.

ERCOT

PublicSlide3

Classic Simplified View of Risk Tolerance(Convex Marginal Propensity to Risk Aversion)

ERCOT PublicSlide4

How to Define ‘Risk Tolerance’ or ‘Risk Appetite’

Definition of ‘Risk Appetite’ (Institute of Risk Management)Risk appetite can be defined as ‘the amount and type of risk that an organization is willing to take in order to meet their strategic objectives’.Organizations will have different risk appetites depending on their sector, culture and objectives. A range of appetites exist for different risks and these may change over time.

ERCOT

PublicSlide5

How to Establish an Organizational Risk Tolerance

Several Methods – Three Common are DescribedQualitative (or ‘gut feel’) Consensus Driven – often done by surveys of key stakeholders – “Do you feel the Organization’s goals and objectives would be compromised if there were a loss of $xxx? $yyy? zzz?”Quantitative Scenario Driven – often the result of some sort of probability analysis (such as a Value-at-Risk or Probability-of-Default calculation) employing basic statistical tools such as a probability distribution or transition matrix and historical pricing, variance, and covariance data in a stochastic or Monte-Carlo simulation with an associated time-weighted probability outcome – “Our analysis shows that there is a 10% chance of a 10-day loss exceeding $xxx and a 1% chance of a 10-day loss exceeding $yyy. Is that acceptable?”

Capital Capacity Driven – A hybrid of a Consensus and Qualitative approach – “Collectively, we the key stakeholders, feel we can only

survive

a 10% chance of a loss of $xxx and a 1% chance of $yyy – do what ever you can to make sure our losses do not exceed these levels!”

ERCOT

PublicSlide6

Considerations in Establishing Risk Tolerances

Key ConsiderationsDifferent stakeholders will have different risk tolerances and risk appetites based upon their Expected Utility of Money which is unique to each stakeholder based upon, among other factors, their risk aversion and opportunity cost and cost of capital. These risk tolerances will likely change amongst stakeholders in a non-uniform manner at each point in the likelihood of outcome and probability of occurrence matrix.

Qualitative Analysis of the Probability of Loss (one measure of exposure of a third-party to the non-performance of another) involves two components - the Exposure at Default which is often not overly difficult to calculate (although recovery rates differ) and Probability of Default which can be very difficult to calculate because it includes both public information available to the third-party as well as private information – which as such is often modeled using default probability matrices developed by organizations such as Moody’s or Standard & Poor’s on a generalized basis and often useful only for rated companies.

ERCOT

PublicSlide7

ERCOT Specific Risk Tolerances Considerations

ERCOT Risk Tolerance Considerations (non-exhaustive)ERCOT has multiple stakeholders, each with different risk appetitesERCOT measures ‘Total Potential Exposure’ (TPE) as a proxy for default exposure, but does not maintain counterparty-specific default probabilities

ERCOT does not have transparency into the capital structure of individual counterparties in order to estimate recovery probabilities

ERCOT’s measure of TPE may overstate market risk due to numerous assumptions and somewhat arbitrary parameters as well as a deliberate design to estimate ‘potential’ and not ‘probable’ exposure

.

ERCOT only sees its TPE exposure; however counterparties have other financial positions which may mitigate exposure (third-party financial hedges, generation outage exposure) or exacerbate exposure (short positions with third-parties, unhedged exposure to commodities such as natural gas needed to operate a plant during a ‘basis squeeze’)

ERCOT lacks the robust analytical tools to calculate portfolio-wide Monte-Carlo or parametric Value-at-Risk or other full quantitative analysis.

ERCOT

PublicSlide8

Next Steps and Questions?

Next Steps?What can CWG/MCWG do to provide F&A with ‘additional guidance on how market risk might be assessed’?Should we differentiate this guidance by ‘normal risk exposure’ (however that is defined) and ‘tail risk exposure’?

How should this guidance be structured and what approaches (if any) should be taken in its development?

When should we target providing this guidance?

Questions?

ERCOT

Public