What are best practices in incorporating macroeconomic variables into stress testing strategies?
Last week, Dr. Dan Rosen, Managing Director of Risk Analytics at S&P Capital IQ, offered his strategy during a webinar entitled “Re-Thinking Scenarios: Integrating Economic Scenarios with Advanced Scenario Analytics to Manage Investment Portfolios.” Rosen utilized a recent Standard & Poor’s Ratings Services economic forecast from April 2015 to illustrate how to derive scenarios that could be used in assessing the riskiness of a multi-asset portfolio of investments.
Dr. Rosen shared his 7 step process with the audience:
Assess risk factors most relevant to the portfolio
- Distill forecast into the relevant risk factors
- Build joint simulation models to combine market factors with the economic factors in the forecast
- Map the model
- Create conditional scenarios to shock the risk factors
- Run simulations
- Analyze results
Additionally, here are a few observations about the creation of scenarios and the use of forecasts:
- Scenario analysis and forecasts serve different purposes. A forecast is a representation of expected market conditions, while a scenario contains the joint realization of financial and economic risk factors at a future point in time.
- Properly crafted scenarios require a deep understanding of both the portfolio, and expected future market conditions.
- Generating long-term forecasts is NOT the job of the risk manager. This is the responsibility of economists. Risk managers adapt the various economic views of future market states and derive from them scenarios that useful is assessing the risk in their portfolios.
- The future cannot be predicted. The goal of scenario analysis is not to predict what will happen. The goal is understand what market conditions could cause undesirable levels of risk or return in the portfolio.
- Scenarios are most useful when they are accompanied by tools that provide a what-if framework for assessing potential actions of the portfolio manager to mitigate the effects of the scenario on the portfolio returns.
To watch Dr. Dan Rosen’s full presentation, you can access it here. Stay tuned next week for an in-depth look at how to create joint simulation models.