USD/ZAR through mid-2026, for a SA-rooted book with USD liabilities

A South African-rooted family office with USD-denominated liabilities has had an unusually quiet year on the FX line. The rand opened 2026 at an average of 16.31 to the dollar, briefly strengthened to 15.73 on 29 January, then weakened to a 2026 high of 17.19 on 30 March before settling back into a 16.0 to 16.8 band through the second quarter (FRED, South African Rand to U.S. Dollar Spot Exchange Rate, DEXSFUS). The monthly averages for 2026 to date: January 16.31, February 16.02, March 16.74, April 16.57, May 16.49, June 16.41, July 16.40 (X-Rates monthly averages, cross-checked against the FRED noon buying rate).

That is not a collapse and not a rally. It is a corridor. For an office that has to fund USD liabilities out of a ZAR base, the work in 2026 has been to write the corridor policy against the shape of the band rather than against any single number on the print. The two-year window matters as much as the year-to-date. Annual averages were 18.33 in 2024 and 17.88 in 2025 (OECD CCUSMA02ZAA618N via FRED, cross-checked against the G.5A Foreign Exchange Rates annual series). The 2026 path has so far taken the rand roughly 1.5 rand stronger on the year, which is a meaningful move at the margin even if it sits inside a band.

The South African Reserve Bank spent 2024 and 2025 cutting the repo rate from the cycle peak of 8.25% down to 6.75%, in six consecutive 25 basis-point moves between September 2024 and November 2025. The MPC then held at 6.75% in January and March 2026, before reversing course with a 25 basis-point hike to 7.00% on 28 May 2026, the first rate increase since May 2023 (South African Reserve Bank, Statement of the Monetary Policy Committee, May 2026). At the 23 July 2026 meeting the committee held at 7.00%, with a 4-2 vote, two members preferring a further 25 basis-point increase.

The rationale is in the July statement. The MPC raised its 2026 inflation forecast to 4.0% (from 4.4% in May), noted June CPI at 5.0%, and pointed to renewed Middle East tensions that took Brent from around $70 a barrel earlier in the month to roughly $90 by mid-July. The committee also explicitly flagged the dollar environment, noting that Fed communications have emphasised the importance of price stability, that rates have shifted up significantly at the short end of the US yield curve, and that in this context the dollar has strengthened against other major currencies. The implication for the rand is direct: a stronger broad dollar, a higher US short end, and a domestic inflation backdrop that has yet to settle at the 3% target.

For a book that runs ZAR-denominated income against USD-denominated liabilities, the practical question is which side of the corridor the policy is sized against. The desk sizes the policy against the band, not the level. The 2026 path of 15.73 on the strong side and 17.19 on the weak side is roughly a 9% range over six months. The two-year range from the 2024 annual average of 18.34 down to the 2026 July spot near 16.40 is closer to 11%. A policy that is written against either side of that range will be wrong the other side.

Three lines anchor a corridor policy that has to work through both sides. The first is the hedge ratio, set in basis points of carry against the ZAR/USD cross, not against an absolute level of the rand. A 50 basis-point carry band around the spot is a unit that holds through a 10% FX move without re-signature. The second is the trigger threshold, set in standard deviations of the trailing two-year monthly average. Monthly USD/ZAR has held in a roughly 15.7 to 18.6 band over the trailing two years; one standard deviation on the monthly series is around 0.7 rand. Writing the trigger at one and a half standard deviations gives a band the policy can hold through without being re-signed.

The third line is the link from the corridor to the underlying liability stack. For an office with USD school fees, USD property carrying costs, or USD-denominated commitments to private fund managers, the matching is rarely dollar-for-dollar. The discipline is to name the gap, name the hedge, name the trigger, and re-read the corridor on a quarterly cycle. If the gap moves outside the corridor for more than two consecutive quarters, the policy says to act. If it does not, the policy says to leave it.

For a SA-rooted office taking on a multi-currency corridor, the work is to size the USD liability stack in writing, name the ZAR funding sources by counterparty and tenor, and match the two through a corridor with a written hedge ratio and trigger. The policy sits in the safe. The quarterly read is the only thing the principal signs. The treasurer signs the trades. The corridor re-reads on a standing cadence against the curve the desk maintains: ZAR/USD monthly average, SARB repo rate, US short end, broad trade-weighted dollar.

On the current set of prints, the corridor is wider than the policy should be. The 16.0 to 16.8 monthly band through 2026 to date is a working range, not the policy. The policy is the two-year band, with the trailing two-year monthly average and standard deviation as the live reference. As long as the rand stays inside one and a half standard deviations of the trailing average, the policy says to leave the hedge alone. As long as the SARB rate path stays roughly within the published Quarterly Projection Model range, the policy says not to chase the rate view. The trades are taken on the corridor edges, not on the prints.

For an office with an actual USD liability book, the 2026 year is shaping up as a year of policy discipline, not of FX view. The rand has done about 10% over two years, SARB has moved 125 basis points, the US short end has moved roughly 170 basis points, and the broad trade-weighted dollar has held in a 117 to 129 band. The views are crowded. The corridor policy is not. The principal signs the corridor once. The treasurer re-reads it every quarter. The next read is September 2026, after the next SARB meeting on the 23rd.