<p>This study examines how the exchange rates of four major foreign currencies–the Australian dollar (AUD), British pound (GBP), Swiss franc (CHF), and Japanese yen (JPY)–influence movements in the U.S. dollar (USD) under different phases of the U.S. economic cycle. Using Markov switching regression on U.S. unemployment rate data, we identify expansion and recession regimes and estimate regime-specific relationships through linear regression, quantile regression (QR), and adaptive LASSO penalized quantile regression (ALQR). The ALQR framework allows for the detection of heterogeneous and tail-dependent linkages that vary across both economic regimes and quantiles of the USD return distribution. The empirical results reveal strong regime asymmetry. During recessions, the CHF exerts the most substantial influence on the USD, consistent with its global safe-haven role, while the sensitivity of the USD to pro-cyclical currencies such as the AUD declines. Conversely, in expansion periods, the AUD and GBP display stronger effects on USD movements, these currencies tend to strengthen when the global economy is booming. The JPY exhibits relatively weak yet asymmetric effects, becoming more influential in lower quantiles, which correspond to downside market conditions. By integrating adaptive penalization into a quantile framework, this study contributes methodologically to the literature on exchange rate connectedness by highlighting regime- and tail-specific dependence patterns often overlooked by mean-based models. The findings carry practical implications for monetary authorities and investors, suggesting that policy and risk management strategies should account for both economic regimes and the conditional dependence structure of major currencies.</p>