Rates first. Interpretation second.
A government par-yield curve records a cross section of yields by maturity. The maturities are labels on an estimated curve, not necessarily prices of bonds with exactly those remaining tenors. Before inferring anything from slope or curvature, the analyst needs the source, the estimation methodology, the fixing time and the reference date.
For the U.S. Treasury, par yields are built from indicative closing market inputs using an established curve-fitting procedure. The Bank of Canada's selected benchmark yields instead refer to identified benchmark issues. These two series are useful but are not identical instruments; direct cross-market comparisons require convention checks.
A reproducible slope
On 10 November 2026, the published U.S. par yields were 4.73% at two years and 4.98% at ten years. Their arithmetic spread was 25 basis points. This is a dated observation, not a prediction of the next monetary-policy decision.
The same desk can report changes measured against the previous published business day. That daily difference is not necessarily equal to the change in any specific tradeable bond because benchmark roll conventions, liquidity and quotation conditions vary.
What duration does—and does not—tell us
For a sufficiently small parallel rate shift, modified duration gives a first-order price approximation. A 100 bp move is not necessarily small, and nonparallel curve shocks require bucketed risk measures rather than a single-duration shortcut.
Convexity, coupon schedules, accrued interest and discounting matter to any actual security valuation. A curve plot alone cannot establish portfolio P&L, implementation cost or expected return.
Data controls
- Keep the official observation date separate from the website retrieval date.
- Never fill a missing tenor with an unmarked model estimate.
- Preserve the original source convention and unit before computing derived spreads.
- State plainly whether values are daily published observations, model-derived analytics or hypothetical scenarios.