Expected and Unexpected Inflation Sensitivity of Gold and the U.S. Dollar in Türkiye: Regime-Dependent Evidence (Bucak İşletme Fakültesi Dergisi)

Summary

• The dataset spans 156 monthly observations from January 2013 to December 2025 across all analyzed financial series.p.5
• Expected inflation averaged 1.55% per month, while realized monthly inflation averaged 1.83% over the sample period.p.5
• Full-sample OLS regressions estimate an expected inflation coefficient of 0.8144 for gold compared to 0.3722 for the U.S. dollar.p.8
• Gold return sensitivity to unexpected inflation shifted from -0.7185 pre-break (106 observations) to 0.4878 post-break (50 observations).p.10
• Rolling 36-month regressions show positive expected inflation coefficients in only 40.5% of windows for gold and 48.8% for the U.S. dollar.p.12

Key insights

1. Gold demonstrates broader inflation sensitivity across return distributions than the U.S. dollar in Türkiye.p.1
2. Unexpected inflation shocks exert little or no systematic impact on either gold or U.S. dollar returns.p.1
3. Structural break analysis identifies a major regime shift clustered around late 2021 across macro-financial variables.p.6
4. Full-sample OLS results indicate gold is consistent with a full expected inflation hedge as H0: β1 = 1 cannot be rejected.p.9
5. Quantile regressions show gold's expected inflation sensitivity reaches 1.7494 at the 0.10 return quantile.p.9
6. The U.S. dollar exhibits expected inflation sensitivity strictly in the lower tail of returns, with a coefficient of 0.6744 at tau = 0.10.p.9
7. Decomposing TRY gold returns reveals its hedging sensitivity relies on combined global gold price and currency channels.p.11
8. Rolling-window estimates confirm that inflation hedging in Türkiye is highly episodic and disrupted during crises.p.12