["International Finance, Volume 29, Issue 1, Page 92-109, Spring 2026. ", "\nABSTRACT\nThis study investigates five German biotechnology firms cross‐listed on XETRA and NASDAQ. By employing high‐frequency data, we estimate both bivariate and trivariate vector error correction models—the latter explicitly accounting for exchange rate dynamics—to assess which market, domestic or U.S., leads in price discovery. The results suggest that XETRA plays a dominant role for larger firms, whereas smaller firms are more influenced by NASDAQ. Simulations show that bivariate models—common in the literature—yield biased results under volatile exchange rate conditions, whereas trivariate models produce more robust estimates. Finally, exchange rate shocks affect NASDAQ and XETRA differently.\n"]