We are medium-term bearish ZAR, but following the sharp sell-off in recent weeks we do not hold active short positions at the moment.

Political risks are likely to be a key driver of the currency, in the near term. It is unclear if Finance Minister Gordhan would remain in his post. Additionally, there is a risk the tensions in the government may decrease policy credibility.

Sovereign credit ratings downgrade risk has also risen.

ZAR experienced a sharp selloff following the latest flare-up in domestic political risks with an increased likelihood that Finance Minister Gordhan may be charged by the police unit.

The sell-off reached about 10% at its peak, matching the correction experienced during the removal of Finance Minister Nene in December 2015.

We held a short ZAR position heading into the event and took profits on August 30.

On the USDZAR front, we recommend positioning long USDZAR (ZAR has been significantly overshooting fundamentals), which makes buying USDZAR vol all the more appealing.

In naked vanilla form, we suggest call spreads at the 2M horizon, optimizing strikes for leverage. In USDZAR, the 1M-2M ATM spread is below average at +0.75, as 1M vols had remained relatively anchored and never softened significantly.

Therefore the premium for owning US elections risk isn't punitive, and the short leg further mitigates the cost of gamma.

The above table explains a ranking where call spreads are ordered in decreasing values of max payout/cost. We find that skews aren’t steep enough vs ATM to allow for a wide range of strikes to be efficient.

In order to ensure more than 50% discount to the outright vanilla, and a max payout/cost higher than 3.5:1, one needs to choose a combination of long 40D vs 25D. The call spread achieves a 55% discount to outright call and a max payout/cost ratio of 3.7:1 (mid values).