Post-midterm, published research is split on US bonds — and the fault line runs through AI.
The near-term tone is yields drifting lower. But the longer view divides sharply. One camp of analysts argues a divided Congress means less fiscal expansion, lighter Treasury supply, and a lower term premium — with falling energy prices and Iran de-escalation adding support. A contrarian thread of recent research pushes back, pointing to deficits still near 6% of GDP, sticky inflation, and tariff risk; one published view sees 10-year yields reaching 5.10%.
On the curve, some researchers expect the 2-year to fall if the Fed drops its hawkish bias, with intermediates favored and a bull-flattening bias. Others argue the long end should sell off on a rising term premium. On inflation-protected bonds, the consensus leans toward keeping duration short.
Credit is where AI matters most. Several published views expect hyperscaler and data-center spreads to widen on heavy issuance; others argue spreads tighten if AI capex slows. The disagreement is real, not semantic.
By the numbers:
• Positioning (CFTC, as of 27 Oct 2026): Speculators (leveraged funds) are net short in all four contracts. US Treasury bond has the smallest net short in a year (COT index 100); US 2-year note (83) and US 5-year note (86) are near the top of their 52-week range; US 10-year note (26) is near the bottom of its 52-week range.
• Federal Reserve policy rate 3.75–4.00%, last move +25 bp on 17 Sep 2026. Next decision 9 Dec 2026.
• United States 10-year yield 4.90%, 2-year 4.58% (U.S. Department of the Treasury, 4 Nov 2026).
Charts: finchiefs.com/data/policy-rates/federal-reserve