Election Follow-up: Probability-Weighting Forecasts
The result of the United States presidential election last week delivered more than just a surprising new Commander in Chief–the markets surprised, too. Initial reactions to the news that Donald Trump was the likely winner at approximately 1:00 EST shocked markets and crushed risk appetite – much as many had expected. Risk was immediately dumped in favor of the usual havens, including gold and CHF, and equity vol jumped more than 20 percent in the early hours. But, as time wore on, prices settled and liquidity improved. As the opening bell rang Friday morning, we were witness to far-reaching reversals as stocks pared losses and pushed on to making strong gains by the end of the day.
Below I illustrate the extent to which markets moved from initial reactions to twelve hours following:
|Asset Class||Clinton Wins||Trump Wins|
|US Bond Prices||-1%||2%|
|Initial Reaction||Change on close (01.00 ET)||Forecast|
|US Bond Prices||0.4%||2%|
|12hrs Later||Change on prev. close (13.00 ET)||12hr Change (% Points)|
|US Bond Prices||-2.3%||-2.7%|
The above table captures the potential nature of event risk outcomes–limit down one moment to unch and then higher within hours. What is crucial to note here is the biggest movement of capital happened sometime after information was known, not as it was known, like many are more familiar.
Below is an example of how clients can assign probabilities to forecasts, which in turn allows for an overall estimated expectation of event outcomes:
Aggregated by Strategy
Being fortunate to predict an initial reaction matters if exposure to prices stops there–the reality can be markedly different.
In times of stress in the markets, not only does volatility increase for individual assets, cross-asset correlations can increase dramatically as well. This results in a “double whammy” for a typical portfolio because the portfolio’s volatility increases due to both effects.
Those responsible for maintaining a margin system often feel that they are drowning in data management issues. In part two of this series we discuss ways to make margin calculations far more efficient and meet the firm’s need for answers in real-time.