Prolepsia forecastNumber

What will Japan's 10-year government bond yield be on 31 December 2027 (Ministry of Finance benchmark), in percent?

As of 17 November 2026, Prolepsia’s median forecast is 3.15, with an 80% chance it falls between 2.04 and 4.23.

3.15median

Prolepsia’s forecast distributionHow likely each value is for: What will Japan's 10-year government bond yield be on 31 December 2027 (Ministry of Finance benchmark), in percent?. Median: 3.15. 80% range: 2.04 to 4.23.
Prolepsia forecast made

How Prolepsia reasons

The report Prolepsia wrote with this forecast.

How forecasts are made · Track record

TL;DR

The forecast for Japan’s end-2027 MOF 10-year government bond yield has a median of 3.15%, a 10th percentile of 2.04%, and a 90th percentile of 4.23%. The central view is a modest further rise, not a repeat of the full repricing away from exceptionally low yields. Continued monetary normalization supports yields, while weak activity and limits to the BOJ’s purchase reductions restrain the increase.

Key evidence

  • The starting yield already reflects substantial normalization. The November 16, 2026 MOF-attributed observation is 2.927%, corroborated by JGB Information. Future policy-rate increases should not be added mechanically to a long-term yield that already incorporates expectations.
  • Policy and inflation still point upward. The BOJ maintained its overnight-rate guideline at 1.25% on October 30, 2026. Its October outlook projects FY2027 inflation excluding fresh food of 2.6%, excluding the proposed consumption-tax reduction.
  • Central-bank withdrawal has a limit, but remains consequential. The purchase plan sets purchases at approximately ¥2 trillion monthly from April 2027. Holdings still decline as bonds mature, leaving private investors to absorb more debt.
  • Supply and growth provide offsets. The revised FY2026 financing plan increased total issuance without increasing scheduled market issuance. Meanwhile, the first July–September GDP estimate showed a 0.1% annualized contraction.

Uncertainties

The largest uncertainty is whether underlying wage and service-price inflation persists despite weak household demand. A tax-driven fall in measured inflation would not establish an equivalent fall in underlying pressure. Fiscal-2027 maturity-specific issuance and private investors’ capacity to absorb BOJ runoff also remain unresolved.

The 2.927% starting observation is corroborated, but was not independently confirmed against the official CSV. The target remains the MOF constant-maturity series, not an individual bond’s yield. Because December 31 is a BOJ holiday, December 30, 2027 is the expected observation date; the final valid observation on or before the deadline controls.

Read the full report

TL;DR

The forecast for Japan’s end-2027 MOF 10-year government bond yield has a median of 3.15%, a 10th percentile of 2.04%, and a 90th percentile of 4.23%. The central view is a modest further rise as monetary normalization and persistent inflation meet weak growth and supply offsets. Use the median as the central planning value and the percentile range to assess exposure to lower yields from weaker demand and higher yields from inflation or fiscal stress.

Context

Japan’s bond market has moved away from the exceptionally low yields associated with yield-curve control. The supplied starting observation is 2.927% on November 16, 2026, corroborated by MOF-attributed market displays. The question is no longer whether yields will leave the old low-rate regime. It is how much further they will rise after substantial normalization has already entered prices.

The outcome is the official nominal 10-year constant-maturity yield, not a coupon, futures yield, or individual bond’s yield. The MOF methodology uses secondary-market prices observed at 3 p.m. Japan Standard Time and semiannual compounding, with publication normally on the following business day. The BOJ holiday calendar makes December 30, 2027 the expected observation date. The actual final valid MOF observation on or before December 31 takes precedence, even if publication occurs after year-end.

Evidence

The historical backbone argues against treating either recent upward drift or recent volatility as permanent. The available analysis of the OECD/FRED monthly benchmark distinguishes two older regimes: annual yield changes ending in 1990–1998 had a standard deviation of about 0.94 percentage point, while those ending in 1999–2012 had a standard deviation of about 0.22 percentage point. Average annual changes were negative in both periods. These are related benchmark data, not the exact daily resolution series (OECD/FRED history).

The more recent history reflects the exit from unusually low rates. In the January 2023–October 2026 sample, overlapping 13-month yield changes ranged from +0.065 to +1.430 percentage points, with a median of +0.565 percentage point. Those overlapping periods are not independent observations, and their positive drift is not a neutral guide to the future. They span a policy transition that has already raised the starting yield (historical benchmark data).

Recent daily data also warn against an overly narrow range. The August 6–November 16, 2026 window contains 67 MOF-attributed yield levels; its daily changes imply annualized volatility of about 0.59 percentage point. The shorter October 1–November 16 window implies about 0.66 percentage point. These are short-window scale checks, not complete measures of future policy or fiscal risk (daily yield history). The inference is that ordinary market fluctuations matter, but shifts in inflation, policy and investor demand justify a broader assessment of uncertainty.

The strongest upward force is the BOJ’s continuing normalization bias. On October 30, 2026, it maintained the overnight-rate guideline at 1.25% by an 8–1 vote; the dissenter preferred 1.50%. This supports further tightening pressure, not a predetermined endpoint for long-term yields (policy decision). A 10-year yield reflects expected short rates over many years plus compensation for holding duration. Prospective policy-rate increases therefore do not translate one-for-one into additional yield increases.

The BOJ’s projections pair subdued growth with persistent inflation. Across FY2026, FY2027 and FY2028, its median real-growth forecasts are 0.7%, 0.7% and 0.8%. Inflation excluding fresh food is projected at 2.5%, 2.6% and 2.0%; excluding both fresh food and energy, the sequence is 2.4%, 2.7% and 2.3%. These are fiscal-year projections, not observations or predictions of the year-end bond yield (October outlook). They support continued normalization while leaving little room for a strong-growth explanation of much higher yields.

The inflation outlook has a crucial qualification. It excludes the proposed temporary food-consumption-tax reduction. Full implementation would mechanically reduce FY2027 inflation excluding fresh food by approximately 1.3 percentage points, and the measure excluding fresh food and energy by approximately 1.4 percentage points (outlook discussion). Those are conditional tax effects, not evidence that wages, services prices or inflation expectations have weakened by the same amount.

Current price signals are mixed. September national inflation was 2.1% excluding fresh food and 2.4% excluding fresh food and energy (national CPI release). October’s preliminary Tokyo release showed the latter measure at 2.6%, alongside rent inflation of 1.9%, even as rice excluding Koshihikari fell 31.3% year over year (Tokyo CPI release). The local release is only one monthly observation. Still, it shows why falling food prices need not imply that domestic inflation pressure has disappeared.

Activity provides a clear counterweight. The first estimate for July–September 2026 showed real GDP down 0.1% at an annualized quarterly rate. Final sales fell 2.1% annualized and real gross domestic income fell 1.6% annualized (Cabinet Office estimate). One preliminary quarter does not establish a recession. It does establish a reason to retain substantial lower-yield risk rather than assume uninterrupted tightening.

Bond purchases create pressure, but the scheduled withdrawal is not open-ended. Monthly purchases are set at ¥2.7 trillion, ¥2.5 trillion, ¥2.3 trillion and ¥2.1 trillion across successive quarters from April 2026 through March 2027, then approximately ¥2 trillion from April 2027. The BOJ retains flexibility to increase purchases during rapid yield rises (purchase plan). Purchases plateau, but holdings still shrink as redemptions exceed purchases. The plan’s conditional holdings path falls from approximately ¥480 trillion in March 2027 to ¥430–440 trillion in March 2028. Private demand must absorb that shift.

Fiscal supply is more nuanced than the borrowing headline. The revised FY2026 plan raised total issuance from ¥180.6920 trillion to ¥183.8055 trillion, while scheduled market issuance remained ¥168.5 trillion (revised financing plan). Relative to revised FY2025, planned annual 10-year issuance remained ¥31.2 trillion. Planned 20-, 30- and 40-year issuance fell from ¥10.2 trillion, ¥8.7 trillion and ¥2.5 trillion to ¥8.4 trillion, ¥7.2 trillion and ¥1.8 trillion, respectively (issuance calendar). These are announced supply offsets, not proof that fiscal-2027 financing will be easy.

Other demand signals deserve less weight. Recent auctions showed different demand conditions across maturities, rather than a uniform refusal to buy government debt. Reports that a new floating-rate bond launch would be postponed were based on unnamed sources and lacked established official confirmation (reported postponement). Neither isolated auction weakness nor an unconfirmed product decision establishes a durable change in demand.

What's non-obvious

Normalization is not the same as an unpriced yield increase. The starting yield already embodies expectations of tighter policy and less BOJ support. Extrapolating the rise from the yield-curve-control era, then adding future rate increases and purchase reductions again, would count the same forces twice. The evidence supports a modest further rise at the center, while preserving a wide range of outcomes.

Three distinctions matter more than the headlines: purchases versus holdings, borrowing authorization versus auction supply, and tax-driven inflation versus underlying inflation. Purchases can stabilize while holdings keep falling. Total borrowing can rise without an immediate increase in scheduled auctions. Measured inflation can fall sharply without an equivalent easing in persistent price pressure. Together, these distinctions explain why neither a simple bond selloff story nor a simple disinflation rally story is sufficient.

Uncertainties

The main macroeconomic gap is the persistence of inflation after temporary food and tax effects are separated out. Wage settlements, services inflation and household spending would help distinguish durable domestic pressure from price-level changes. Final enactment and implementation of the proposed tax reduction were not established, so its estimated effects remain conditional.

The main market gap is fiscal-2027 maturity-specific issuance and the strength of private demand as BOJ holdings decline. The available plan does not cover the full target year. No contemporaneous decomposition of expected short rates and term premium was established, and there is no verified basis for assuming a large insurer or pension repatriation wave.

The starting observation is corroborated by MOF-data reproductions but was not independently confirmed against the official CSV. Historical proxies also differ from the resolution series. These limitations reduce confidence in fine distinctions, not in the target definition. Persistent inflation, currency weakness or fiscal repricing would push toward higher outcomes; weak demand, disinflation or a global downturn would push toward lower ones. The saved percentile range captures substantial uncertainty in both directions and is not a hard bound.

Forecast history

How Prolepsia’s forecast has movedOne forecast so far: 3.15 on 17 Nov 2026.
DateForecast
Median 3.15, 80% range 2.04 to 4.23

Sources Prolepsia read

  1. 概況 | 日本国債インフォメーションsaiken.net
  2. Interest Rate (Q & A): Ministry of Financewww.mof.go.jp
  3. Holiday Schedule of the Bank : 日本銀行 Bank of Japanwww.boj.or.jp
  4. Interest Rates: Long-Term Government Bond Yields: 10-Year: Main (Including Benchmark) for Japan (IRLTLT01JPM156N) | FRED | St. Louis Fedfred.stlouisfed.org
  5. 統計局ホームページ/消費者物価指数(CPI) 全国(最新の月次結果の概要)www.stat.go.jp
  6. Quarterly Estimates of GDP for July - September 2026 (First Preliminary Estimates) summarywww.esri.cao.go.jp
  7. Japan to postpone launch of new floating-rate government bonds, sources say | MarketScreenerwww.marketscreener.com
  8. US–Japan government bond yield spread | Tokubakutokubaku.com
  9. 主要年限レート(長期金利等) | ヒストリカルデータ | 日本相互証券www.bb.jbts.co.jp

The question

This question asks for the yield, expressed as an annual percentage rate, on Japanese government bonds (JGBs) with a constant remaining maturity of 10 years at the end of calendar year 2027, using the official benchmark yield series published by Japan's Ministry of Finance (MOF). The forecast is made as of November 17, 2026, with a target date of December 31, 2027. The relevant measure is the MOF's 10-year constant-maturity government bond yield, not the coupon rate on newly issued 10-year bonds, an individual bond's yield to maturity, or the yield on a 10-year bond futures contract. The MOF calculates its daily constant-maturity yields using secondary-market prices of fixed-rate JGBs, with yields expressed on a semiannual-compounding basis. The underlying market observations are taken at 3:00 p.m. Japan Standard Time, and the resulting rates are normally published on the following business day. Source: Ministry of Finance, Interest Rate Q&A (https://www.mof.go.jp/english/policy/jgbs/reference/interest_rate/qa.htm). As background, Japan's 10-year government bond yield has risen substantially from the exceptionally low levels associated with the Bank of Japan's earlier yield-curve-control policies. The MOF-derived 10-year benchmark yield stood at approximately 2.927% on November 16, 2026, providing a recent reference point for the forecast. Future yields will depend on factors including Bank of Japan monetary policy, inflation expectations, fiscal developments, government bond supply and investor demand. An important calendar consideration is that December 31, 2027, falls on a Friday and is a Japanese banking holiday. The Bank of Japan's official 2027 holiday schedule confirms that December 31 is a closure date. Consequently, the forecast will use the final available MOF 10-year yield observation dated on or before December 31, 2027, which is expected to be December 30, 2027. Sources: Bank of Japan, Holiday Schedule (https://www.boj.or.jp/en/about/outline/holi.htm); Ministry of Finance, Interest Rate (https://www.mof.go.jp/english/policy/jgbs/reference/interest_rate/index.htm).

How it resolves

The question will resolve to the numerical value of the 10-year constant-maturity Japanese government bond yield published by Japan's Ministry of Finance in its official Interest Rate dataset for the final available observation date on or before December 31, 2027. The primary resolution source is the Ministry of Finance's official Interest Rate page and its associated current or historical CSV datasets (https://www.mof.go.jp/english/policy/jgbs/reference/interest_rate/index.htm). The relevant data point is the value in the 10-year maturity column, expressed in percent per annum, for the last observation date in calendar year 2027. Because December 31, 2027, is a Japanese banking holiday, the expected observation date is Thursday, December 30, 2027. If the MOF publishes a valid observation specifically dated December 31, 2027, that observation will take precedence. Otherwise, the latest valid observation dated before December 31, 2027, will be used. The reported yield will be taken directly from the official dataset without converting compounding conventions, averaging observations, or substituting an individual bond's yield. The full numerical precision published by the MOF will be retained, normally three decimal places in percentage terms. The question will be resolved once the relevant MOF data become available, allowing for publication delays associated with the Japanese New Year holiday period. If the Ministry of Finance subsequently corrects the relevant observation before resolution is finalized, the corrected official value will be used. If the MOF permanently discontinues its existing series before the target date but publishes a directly comparable successor series, that successor will be used provided it retains the same underlying concept of a nominal 10-year constant-maturity JGB yield. If no directly comparable MOF observation or official successor series is available for any date in December 2027, the question will be annulled rather than resolved using an unrelated market indicator or an independently constructed estimate. No interpolation across missing dates will be performed.

Fine print

The forecast concerns a yield level, not a change in yield, bond price, coupon rate, or total investment return. Values are expressed as annual percentage rates, so an outcome of 3.250 represents a yield of 3.250%, not 0.03250%. A change of 0.100 in the reported value corresponds to 10 basis points. The forecast range of 0% to 6% is intended to cover a broad set of plausible outcomes rather than impose hard limits. Negative yields and yields above 6% remain theoretically possible. The observation date is determined according to the Japanese calendar and Japan Standard Time (UTC+9). The resolution uses the final available observation on or before December 31, 2027, rather than the first observation published in 2028. A dataset published in January 2028 containing observations from December 2027 remains eligible. The benchmark is the MOF's constant-maturity nominal yield, which is constructed from market information and should not be confused with the yield of a particular actively traded benchmark bond. Temporary trading suspensions or unexpected market closures will not automatically annul the question; the final available qualifying MOF observation in December 2027 will determine the outcome. Subsequent routine historical revisions made after the question has been resolved will not reopen resolution.