Prolepsia forecastYes / no

Will the US 10-year Treasury yield be above 4.5% on 31 December 2027?

As of 14 November 2026, Prolepsia puts the chance of yes at 69%.

69%chance of yes

Yes 69%No 31%
Prolepsia forecast made

How Prolepsia reasons

The report Prolepsia wrote with this forecast.

How forecasts are made · Track record

TL;DR

The forecast is 69.2% Yes and 30.8% No that the official nominal 10-year Treasury yield will be strictly above 4.50% on the applicable year-end 2027 observation. The starting yield and current policy backdrop favor Yes, but a decline large enough to produce No is well within historical experience.

Key evidence

  • A meaningful starting cushion. Treasury reported a 10-year yield of 5.15% on November 13, 2026, putting it 65 basis points above the threshold. A decline to exactly 4.50% would resolve No (Treasury).
  • Policy and inflation favor persistence. The October 28 Federal Reserve statement held the policy target at 3.75–4.00%, with two dissenters preferring an increase (FOMC). October consumer prices rose 0.4% month over month, with core prices up 0.3%, both seasonally adjusted (BLS).
  • History limits confidence. Across 35 comparable November-to-following-December changes spanning starting years 1990–2024, the mean change was −0.12314 percentage points and the standard deviation was 1.03077 percentage points. That dispersion exceeds the present cushion (FRED historical series).
  • There is a credible opposing forecast. TD Economics’ November 13 publication contains an October-vintage forecast of 4.35% for the end of 2027, below the resolution threshold. Its horizon is relevant, though it predates the latest inflation release (TD Economics).

Uncertainties

The largest risk is that high real borrowing costs weaken growth, bringing lower expected policy rates and lower long-term yields. Disinflation or a decline in the compensation investors demand for holding long bonds could also produce No without a recession.

Conversely, persistent inflation and sustained government borrowing pressure favor Yes. Historical comparisons span different monetary regimes and include overlapping periods, so they establish the scale of uncertainty better than a precise event probability.

Resolution depends on the published year-end observation, not any earlier crossing. Exactly 4.50% is No; 4.51% or higher is Yes. If no observation is published for December 31, the specified preceding-business-day fallback applies.

Read the full report

TL;DR

The forecast is 69.2% Yes and 30.8% No that the official nominal 10-year Treasury yield will exceed 4.50% on the applicable year-end 2027 observation. The starting yield and current inflation-policy backdrop favor Yes. The key takeaway is that the present cushion supports persistence, not certainty: historical yield swings are larger than the decline needed for No.

Context

This forecast is dated November 14, 2026, roughly 13.5 months before the target. The latest Treasury observation is 5.15% on November 13, leaving a 65-basis-point cushion above 4.50%. The same curve reports 4.88% at two years and 5.47% at thirty years, so elevated yields extend across maturities rather than being confined to the ten-year point (Treasury yield curve).

The outcome uses Treasury’s published nominal ten-year constant-maturity par yield, not a particular bond, an intraday quote, or an annual average. A published 4.51% or higher means Yes; exactly 4.50% means No. The relevant observation is December 31, 2027, or the specified latest preceding business day if that date has no publication (resolution dataset).

Evidence

The historical backbone is simple: long yields move enough over this horizon to erase the starting cushion. Across 35 November-to-following-December changes, using starting years 1990–2024, the average change was −0.12314 percentage points, the median was zero, and the sample standard deviation was 1.03077 percentage points. Applied to the current starting yield, 24 of those 35 changes would leave the yield above the publication-adjusted threshold (historical DGS10 series).

The broader historical comparisons tell the same story, with a warning about mean reversion. These are calculations from month-end observations, not independent trials or direct forecasts of this event (FRED).

Historical comparison Completed changes Mean change, percentage points Standard deviation, percentage points
Annual November starts, 1990–2024; 13 months 35 −0.123143 1.030767
Annual October starts, 1990–2024; 14 months 35 −0.179429 1.063888
All monthly starts from 1990; 13 months 429 −0.1232 0.9223
All monthly starts from 1990; 14 months 428 −0.1325 0.9528
All monthly starts from 2000; 14 months 308 −0.0640 0.8798
Monthly starts from 2000 at yields of 4.50–6.00%; 14 months 55 −0.6149 0.5433

The broad samples support persistence with moderate normalization. The starting-level-conditioned sample shows a stronger tendency for yields to fall. But it is concentrated in relatively few episodes, and overlapping windows reduce the effective amount of independent evidence. The reliable lesson is the scale of possible movement, not a mechanically transferable historical success rate.

Professional forecast errors reinforce that point. Philadelphia Fed statistics published August 22, 2025 report a four-quarter-ahead ten-year-yield forecast RMSE of 1.04 percentage points over 1993 Q1–2023 Q1, with 121 observations. These are errors for quarterly averages, not single-date outcomes, so the figure is an uncertainty check rather than an exact predictive standard deviation (Philadelphia Fed error statistics).

Current policy evidence favors elevated yields. On October 28, the FOMC maintained its target at 3.75–4.00%, described activity as strong and inflation as elevated, and recorded two dissents favoring an increase. That does not support treating a large easing cycle as the default (FOMC statement). October CPI then rose 0.4% month over month, while core CPI rose 0.3%, its third consecutive such monthly increase. The usual October annual comparison was unavailable because the corresponding October 2025 observations were missing (BLS release).

The Federal Reserve’s projections nevertheless allow substantial disinflation. September’s median PCE inflation projection falls from 3.7% in 2026 to 2.3% in 2027; projected 2027 core PCE inflation is 2.5%. The median end-2027 policy rate is 4.1%, against a longer-run assessment of 3.2%. These are conditional policymaker projections, not ten-year Treasury forecasts. They support normalization without establishing that the resolving yield will fall below the threshold (September projections).

The composition of the yield increase matters. The available observations show that the rise in nominal yields has tracked higher real yields, while the simple nominal-minus-real spread barely moved. The historical endpoints are reported in the Treasury comparison; the matching November 12 observation is also documented in the Federal Reserve’s daily rates release (Treasury comparison; Federal Reserve daily rates).

Observation date, 2026 Nominal ten-year yield Real ten-year yield Simple difference
January 2 4.19% 1.94% 2.25%
November 12 5.10% 2.86% 2.24%
November 13 5.15% 2.91% 2.24%

This spread is not a pure inflation forecast. Still, it argues against assuming that lower inflation alone will reverse the entire rise. It also identifies a downside channel: real yields can fall if growth weakens or investors accept lower compensation for holding long bonds. The Kim–Wright model estimated a ten-year zero-coupon term premium of 0.8738 percentage points on November 6. That estimate is model-dependent and is not an observed component that can simply be subtracted from the resolving par yield (term-premium series).

Borrowing and demand evidence adds support, but not proof, for persistence. Treasury advisory minutes anticipate coupon-size increases in the second half of fiscal 2027, concentrated in the two- to ten-year sector. This is prospective supply pressure, not an immediate issuance shock (Treasury advisory minutes). The November 10 ten-year auction cleared at 5.020%, with bid-to-cover of 2.46. One auction does not establish a durable buyers’ strike (auction results). Meanwhile, initial jobless claims of 200,000 and a four-week average of 195,250 argue against an already-developing broad layoff shock, not against a later downturn (Labor Department).

The clearest opposing evidence is TD Economics’ end-2027 yield forecast of 4.35%. It appears in a November 13 publication but is an October-vintage forecast. Its end-period horizon closely matches the question; its older information set limits its weight against newer inflation and market evidence (TD Economics). It remains a substantive reason to take the No outcome seriously.

What's non-obvious

The important distinction is between lower inflation and lower nominal yields. The observed yield increase was almost entirely matched by higher real yields, not a widening nominal-real spread. My reading is that the decisive question is whether those real yields persist. Strong growth and borrowing demand can sustain them; the economic drag from high real borrowing costs can undermine them.

A second distinction is between the path and the endpoint. Yields can decline during 2027 and still resolve Yes, or remain elevated for most of the year and resolve No. A year-end forecast below the threshold therefore bears more directly on this question than an annual-average forecast. The current cushion is useful evidence, but it is neither a floor nor a requirement that yields stay above 4.50% throughout the year.

Uncertainties

The main unknown is whether disinflation arrives alongside continued growth or a material slowdown. Both can lower yields, but a downturn creates a stronger route to lower expected policy rates and real yields. More persistent inflation, stronger investment demand, or higher required risk compensation would work in the opposite direction.

The evidence is thinner on the durability of term premia and the market’s capacity to absorb future issuance. A model estimate and one auction cannot settle those questions. Further inflation releases, labor-market data, borrowing plans, and repeated auction results would help distinguish persistent pressure from a reversible repricing.

There is also a calendar qualification. December 31, 2027 is the observed federal New Year’s holiday for January 1, 2028, but that alone does not establish whether Treasury will publish an observation (federal holiday calendar). The specified preceding-business-day rule handles a missing publication. Corrections within the question’s 30-calendar-day window also apply. These details determine which official print counts; they do not alter the economic case behind the forecast.

Forecast history

How Prolepsia’s forecast has movedOne forecast so far: 69% on 14 Nov 2026.
DateForecast
69% chance of yes

Sources Prolepsia read

  1. Daily Treasury Rates | U.S. Department of the Treasuryhome.treasury.gov
  2. Market Yield on U.S. Treasury Securities at 10-Year Constant Maturity, Quoted on an Investment Basis (DGS10) | FRED | St. Louis Fedfred.stlouisfed.org
  3. Philadelphia Fed error statisticswww.philadelphiafed.org
  4. Federal Reserve Board - Federal Reserve issues FOMC statementwww.federalreserve.gov
  5. Consumer Price Index Summary - 2026 M10 Resultswww.bls.gov
  6. Source 1www.treasurydirect.gov
  7. News Releasewww.dol.gov
  8. TD Economicseconomics.td.com
  9. Federal Holidayswww.opm.gov

The question

This question asks whether the nominal yield on the 10-year United States Treasury security will exceed 4.50% on December 31, 2027. The forecast is made as of November 14, 2026, approximately 13.5 months before the target date. The relevant measure is the 10-year constant-maturity Treasury (CMT) yield published by the U.S. Department of the Treasury in its Daily Treasury Par Yield Curve Rates dataset. This is a standardized yield for a hypothetical Treasury security with exactly 10 years remaining to maturity, rather than the yield of a particular outstanding Treasury note. As of November 13, 2026, the 10-year Treasury yield was approximately 5.15%, above the 4.50% threshold. Treasury yields have risen amid concerns about persistent inflation, the outlook for Federal Reserve monetary policy, and government borrowing costs. The question therefore concerns whether the 10-year yield will remain above 4.50%, or return above that threshold after any intervening decline, by the final trading day of 2027. The outcome will be determined using the official daily yield observation for December 31, 2027, rather than an intraday maximum, annual average, or yield observed at any earlier point in the year. Background sources: U.S. Department of the Treasury, Daily Treasury Par Yield Curve Rates (https://home.treasury.gov/resource-center/data-chart-center/interest-rates/TextView?type=daily_treasury_yield_curve); U.S. Department of the Treasury, Interest Rate Statistics (https://home.treasury.gov/policy-issues/financing-the-government/interest-rate-statistics).

How it resolves

The question resolves YES if the U.S. Department of the Treasury's official Daily Treasury Par Yield Curve Rates dataset reports a 10-year nominal Treasury yield strictly greater than 4.50% for December 31, 2027. For example, a published yield of 4.51% or higher resolves YES. The question resolves NO if the official yield is 4.50% or lower, including exactly 4.50%. The primary resolution source is the U.S. Department of the Treasury's Daily Treasury Par Yield Curve Rates dataset, specifically the column labeled '10 YR' for December 31, 2027, available at https://home.treasury.gov/resource-center/data-chart-center/interest-rates/TextView?type=daily_treasury_yield_curve. The Treasury's published daily observation, rather than a particular intraday market quotation, determines the outcome. December 31, 2027, is a Friday and is expected to be a regular Treasury market trading day. If the Treasury does not publish an observation for that date because of an unexpected market closure, holiday observance, or other interruption, the observation from the latest preceding business day in December 2027 for which an official 10-year yield is available will be used. If the Treasury's historical dataset is temporarily unavailable, resolution will be delayed until the information becomes accessible. If the Treasury permanently discontinues the relevant series, the Federal Reserve Bank of St. Louis FRED series DGS10, which tracks the daily 10-year Treasury constant-maturity yield, will be used as the fallback source (https://fred.stlouisfed.org/series/DGS10). If neither the Treasury nor FRED provides a usable observation for the target date or the applicable preceding business day, and no authoritative equivalent of the same nominal 10-year constant-maturity yield can be established, the question will be annulled rather than resolved based on an estimate or an alternative financial instrument.

Fine print

The 4.50% threshold is strict: a value of exactly 4.50% resolves NO. The yield is expressed as an annualized percentage, not a bond price, inflation-adjusted real yield, or yield spread. The official nominal 10-year constant-maturity yield is used, not the yield on the most recently issued 10-year Treasury note or any specific outstanding bond. The comparison uses the precision reported in the official Treasury dataset, ordinarily two decimal places, without attempting to infer additional unpublished precision. The question concerns only the yield on the specified date; yields exceeding 4.50% at other times during 2027 do not independently satisfy the condition. Any corrections to the official observation published within 30 calendar days after December 31, 2027, will be incorporated before final resolution. Subsequent revisions will not change the resolved outcome. If the official Treasury series is renamed or its publication format changes without a material change in the underlying yield measure, the successor series will be accepted.