花旗发布2026年美国利率展望。报告认为短期内美债不宜逢低买入，因美联储反应函数不确定性高：10月按兵不动难降低12月加息溢价，核心PCE在11-12月季节性偏弱、1-2月偏强，增长仍强且金融条件紧缩程度存疑。中期看，3至5年期曲线段存在价值，市场未充分反映核心PCE的下修，且受同比基数效应影响，美联储难以在远期曲线之上按季度节奏持续加息。报告预计11月再融资会议或削减甚至取消20年期国债发行，建议以10s20s曲线陡峭化做多20年期利差；维持做空2年期盈亏平衡通胀并以汽油多头对冲，认为市场高估2027年通胀风险；同时做多久期互换利差，预计监管可降低约5000亿美元准备金需求，相当于2027/2028年流动性注入。2026年底基准情形收益率预测：2年期4.80%、5年期4.85%、10年期5.00%、30年期5.30%。
Citi's 2026 US Rates Outlook argues it is too early to buy the dip in USTs in the short term but that there is value in the medium term. There is no obvious near-term catalyst for a UST rally given high uncertainty around the Fed's reaction function: a hold at the October meeting is unlikely to reduce the risk premium for a December hike, core PCE is seasonally weak in Nov/Dec but strong in Jan/Feb, growth remains strong and the tightness of financial conditions is still debatable. In the medium term there is value in the 3y-5y part of the curve, as the market underappreciates downward core PCE revisions and it will be challenging for the Fed to hike beyond forwards on a quarterly cadence due to YoY base effects. Citi sees a strong risk that Treasury reduces/cancels the 20y bond at the November refunding and positions long 20y spreads against 10y spreads (10s20s steepeners). It remains short 2y breakevens hedged via gasoline longs, is long belly swap spreads, and sees scope for ~$500bn