The Impact of Excluding Nonfinancial Exposure on the Usefulness of Foreign Exchange Sensitivity-Analysis Risk Disclosures

Abstract

This paper investigates whether the exclusion of nonfinancial items from the scope of sensitivity-analysis disclosures hinders the usefulness of this quantitative information for predicting exchange rate exposure. We focus on foreign exchange (FX) risk because a large volume of nonderivatives FX exposures is in transactions that are nonfinancial in nature. We review the sensitivity-analysis disclosures of a sample of FX derivatives and find that more than half of these firms chose to exclude nonfinancial exposure from their quantitative disclosures. Hence, the reported sensitivity-analysis loss estimates include mainly the exposures of the derivatives positions and do not reflect the net exposure to FX risk. We further find that entity-level sensitivity, but not derivatives-level sensitivity, exhibits incremental predictive power for exchange rate exposure. This is consistent with the notion that exclusion hinders the usefulness of the sensitivity-analysis disclosures for predicting exchange rate exposure. These results are obtained after controlling for potential sample selectivity bias, traditional risk measures, lagged exchange rate exposure, and other derivatives-related disclosures.