In Japan’s JGB market in September 2026, the 2-year yield rose 20.9 bp from the previous month-end. By contrast, the 30-year yield rose just 0.6 bp, making the short-end-led bear flattening pronounced. The 10-year yield reached 3.057% at month-end, entering the 3% range, while the provisional real 10-year yield rose further to 1.157%. Although the ultra-long-term spread narrowed, the broad rise in interest rates will feed through to the fiscal position with a lag, adding pressure to debt-service costs. Tax revenue data are not yet available, so this issue focuses on movements in rates, prices, and economic activity.
Yield Curve Dynamics: The Driver of Rising Yields Shifts from Medium and Long to Short Maturities
Yields rose across all maturities in September, but the shorter the maturity, the larger the increase, resulting in a bear flattening of the curve. Changes in key maturities, based on month-end data released by the MOF (Ministry of Finance), are as follows.
| Maturity | Month-end (%) | Change from prior month-end (bp) | Change from 3 months earlier (bp) | Change from 6 months earlier (bp) |
|---|---|---|---|---|
| 2Y | 1.952 | +20.9 | +57.0 | +57.7 |
| 5Y | 2.399 | +16.6 | +46.2 | +59.5 |
| 10Y | 3.057 | +11.4 | +36.7 | +69.1 |
| 20Y | 3.877 | +6.2 | +25.2 | +59.3 |
| 30Y | 4.098 | +0.6 | +22.5 | +45.8 |
Over the six months, yields in the 5- to 20-year sector rose by around 60–70 bp, indicating a rise led by medium and long maturities. Over the three months and the past month, the 2-year yield recorded the largest increase. This marks a shift in the driver of rising yields from medium and long maturities to the short end.
In terms of curve shape, the curve remains upward-sloping through the 25-year maturity (4.131%). The 30-year (4.098%) and 40-year (4.099%) yields are slightly below the 25-year yield, leaving the ultra-long end flat to lower. The overall level of the curve continues to shift upward. At the same time, its shape is flattening in the short-to-medium sector and flattening at the ultra-long end; these developments should be considered separately.
The fiscal implication is that the cost of new issuance and refinancing is rising fastest for short- and medium-term bonds. Because short-term bonds are refinanced more frequently, a continuation of this trend would accelerate the pass-through to debt-service costs.
Term Spreads: Two Spreads Narrow Simultaneously
Both the 10Y–2Y and 30Y–10Y spreads narrowed by around 10 bp in September. However, the two declines reflect movements in different parts of the curve.
| Month-end | 10Y–2Y (%pt) | 30Y–10Y (%pt) | 40Y–10Y (%pt) |
|---|---|---|---|
| 2026-06 | 1.308 | 1.183 | 1.102 |
| 2026-07 | 1.294 | 1.181 | 1.166 |
| 2026-08 | 1.200 | 1.149 | 1.151 |
| 2026-09 | 1.105 | 1.041 | 1.042 |
The 10Y–2Y spread has narrowed for three consecutive months since its June peak, contracting by a cumulative 20.3 bp. The 30Y–10Y spread is at its lowest level in the 2026 data period.
10Y–2Y: Two Interpretations and a Check Against Economic Indicators
The 10Y–2Y spread narrowed because the 2-year yield rose faster than the 10-year yield. This move has two possible interpretations: one is that short- and medium-term bonds are pricing in higher policy rates ahead; the other is that it signals an economic slowdown. The available data alone do not allow us to determine which factor is driving the move.
As discussed below, however, both the July Indexes of Business Conditions and the June Tankan survey show recent improvement. Thus, current real-economy indicators do not confirm signs of a slowdown. On the other hand, Tankan business expectations are below current conditions, so the slowdown interpretation cannot be ruled out entirely.
30Y–10Y: Narrowing as the 10-Year Catches Up
The 30Y–10Y spread narrowed as the 10-year yield caught up with ultra-long-term yields. Over the past six months, the 10-year yield rose 69.1 bp, compared with 45.8 bp for the 30-year. After moving into the 4% range, the 30-year yield was nearly flat from the previous month-end.
This spread is often considered an indicator of fiscal confidence at the ultra-long end, but it does not directly measure the fiscal premium. Possible factors behind the narrowing include supply and demand for ultra-long-term bonds, the pass-through of rising yields to the 10-year sector, and changes in the assessment of fiscal risk; the data do not allow these factors to be disentangled. At least this month, there was no evidence that ultra-long-term yields were rising faster than the 10-year yield.
Real Interest Rates: Long-Term Rates Remain Positive, Short-Term Rates Are Still Negative
The provisional real 10-year yield returned to 1.157%, matching May’s 2026 high. In the long-term sector, the absence of financial repression through negative real interest rates continues.
| Month | Nominal 10Y (%) | CPI year-on-year (%) | Real 10Y (%) |
|---|---|---|---|
| 2026-06 | 2.690 | 1.6 | 1.090 |
| 2026-07 | 2.801 | 1.9 | 0.901 |
| 2026-08 | 2.943 | 1.9 | 1.043 |
| 2026-09 | 3.057 | 1.9※ | 1.157※ |
※September figures are provisional, using the August year-on-year CPI rate because September CPI has not yet been published. They will be revised once September CPI is released.
According to MIC (Ministry of Internal Affairs and Communications) statistics, August CPI was 1.9% for the all-items index, 1.7% for core, and 1.9% for core-core. With inflation nearly flat, nominal yields are rising; the increase in real yields since July can therefore be explained almost entirely by higher nominal yields.
By contrast, the real short-term interest rate—calculated by subtracting the August year-on-year CPI rate from the call rate (1.227% in September)—is −0.673% (provisional). It remains in negative territory.
- Government: Real borrowing costs on new long-term debt are positive, limiting the erosion of debt through inflation
- Savers: Holders of long-term bonds can earn positive real returns, while short-term market rates remain negative in real terms
- Businesses: Real financing costs for long-term capital investment are rising, raising the hurdle for investment decisions
The configuration of positive long-term real rates and negative short-term real rates shows that the gap between nominal short- and long-term rates spans the rate of CPI inflation. How this configuration changes will depend on movements in both policy rates and prices.
Implications of BOJ Policy: Short- and Medium-Term Yields Significantly Exceed the Call Rate
According to BOJ (Bank of Japan) statistics, the call rate was 1.227% in September. At month-end, the 1-year yield was 1.684% and the 2-year yield was 1.952%, respectively 45.7 bp and 72.5 bp above the call rate.
These gaps can be interpreted both as the market pricing in higher policy rates over the next one to two years and as including an additional term premium. Under either interpretation, short- and medium-term yields are well above the policy rate, and the cost of issuing new short- and medium-term bonds has already risen substantially.
The ownership structure has changed. In the scorecard of the Fiscal Health Check (August 2026 issue), based on flow-of-funds data at market value, the BOJ’s share of holdings fell from 51.7% in 2025 Q1 to 47.9% in 2026 Q1. Over the same period, foreign ownership rose from 6.3% to 8.1%, and domestic banks’ share rose from 12.0% to 13.2%. Since market-value shares also reflect price movements, these changes do not allow us to identify the BOJ’s purchasing trends or the causes of this month’s yield movements.
The monetary base is approximately ¥543 trillion. The scorecard estimates the BOJ’s interest expense on reserves at an annualized ¥4.2 trillion (August 2026, approximate). Higher policy rates increase this cost and may indirectly affect the fiscal position through the BOJ’s earnings and remittances to the government.
Consistency with the Real Economy: Economic Indicators Continue to Improve
The Cabinet Office’s Indexes of Business Conditions indicate that the economy continued to improve as of July.
| Month | Leading index | Coincident index | Lagging index |
|---|---|---|---|
| 2026-04 | 115.9 | 118.3 | 111.6 |
| 2026-05 | 116.2 | 118.3 | 111.2 |
| 2026-06 | 116.2 | 118.9 | 111.8 |
| 2026-07 | 117.7 | 120.6 | 113.0 |
In July, the leading index rose 1.5 points month on month and the coincident index rose 1.7 points, with both gaining momentum. If supported by economic activity, rising yields can be offset in their effect on debt dynamics insofar as they accompany higher nominal growth.
However, the index data run through July, while the yield data run through September, leaving a two-month gap. Whether the leading index continues to rise in August will be the next point to watch in assessing consistency between higher short- and medium-term yields and the real economy.
Business Sentiment: Strong Current Conditions, Cautious Outlook
In the BOJ Tankan survey for June 2026, the business conditions DI for large manufacturers improved by 5 points from the previous survey to 22. The survey also confirms that businesses expect conditions to deteriorate going forward.
| Survey | Large manufacturers | (Outlook) | Large nonmanufacturers | (Outlook) | Small and medium manufacturers |
|---|---|---|---|---|---|
| 2025-Q4 | 15 | 12 | 34 | 28 | 6 |
| 2026-Q1 | 17 | 15 | 36 | 28 | 7 |
| 2026-Q2 | 22 | 14 | 37 | 29 | 9 |
The outlook for large manufacturers is 14, leaving a gap of −8 from current conditions, wider than the −2 gap in the previous survey. Among nonmanufacturers, the outlook is also substantially below current conditions, at 29 versus 37. The assumed exchange rate among large manufacturers is ¥151.55 per dollar, with continued revisions toward a weaker yen.
Strong current business conditions are supportive of corporate tax revenue. However, actual tax revenue data are not available in this issue. With cautious expectations coinciding with rising yields, whether current DI readings converge toward the outlook in the next survey in September will be a turning point for both the economy and the fiscal position.
Implications for the Current Account: Trade Deficit Widens, Sustainability Conditions Hold Through July
MOF trade statistics show that the trade balance has been in deficit for four consecutive months, with the deficit widening.
| Month | Exports (¥100 million) | Imports (¥100 million) | Trade balance (¥100 million) |
|---|---|---|---|
| 2026-05 | 94,991 | 98,942 | −3,951 |
| 2026-06 | 109,265 | 113,388 | −4,123 |
| 2026-07 | 115,094 | 121,543 | −6,449 |
| 2026-08 | 100,433 | 111,552 | −11,119 |
The August deficit was the second-largest in the period shown, after January (−¥1,165.8 billion). The deficit widened because exports fell more than imports month on month. The cumulative deficit from May through August was −¥2,564.2 billion.
The scorecard assesses Japan-specific fiscal sustainability conditions as follows:
- Current account surplus: The trailing 12-month surplus was ¥35.8 trillion (August 2025–July 2026), up from ¥29.1 trillion in the same period a year earlier. The current account remained in surplus over a period that included trade deficits through July
- Domestic ownership: The combined share held by the BOJ, domestic banks, and insurance and pension funds was 79.5% (2026 Q1, market value), preserving the domestically held structure
- Signs of change: Foreign ownership rose to 8.1%, increasing the relative influence of foreign investors’ activity on JGB supply and demand
Both conditions hold for the periods for which data are available. The wider August trade deficit is not included in the current account aggregation above. Its impact on the current account from August onward will need to be assessed using future balance-of-payments statistics.
Outlook: Debt-Service Costs and the Path of the r−g Differential
This month’s rise in yields will push up interest costs with a lag through refinancing. The Fiscal Health Check scorecard shows general-account interest payments as a share of tax revenue rising from 10.8% in FY2025 to 15.6% in FY2026. The general-government r−g differential, based on IMF data, also narrowed in negative terms, from −5.22 pt in 2023 to −2.90 pt in 2024. These figures, however, relate to periods before this month, and the impact of September’s yield rise on r−g cannot yet be confirmed.
The speed at which market rates pass through to the effective interest rate is determined by the maturity structure of the debt. The average remaining maturity is 9.3 years for the debt stock, but the flow has shortened to 7.0 years (FY2026, JGB issuance plan). Gross financing needs are 26.7% of GDP (FY2026, general account). If short- and medium-term bonds continue to lead the rise in yields, refinancing will push up the effective interest rate.
The government’s central objective is a steady decline in the ratio of general government debt to GDP, with the primary balance managed over multiple years. Under this framework, achieving the target depends on how far the negative r−g differential can be maintained. The key question is whether sustained nominal growth can offset the rise in the effective interest rate.
Points to monitor from next month onward include:
- 2Y–call rate gap (72.5 bp this month): Whether the gap between short- and medium-term bonds and the policy rate widens or narrows
- 30Y–10Y spread (1.041%pt this month): Whether it falls below 1%pt or begins to widen again
- September CPI: Whether the provisional real 10-year yield of 1.157% holds in the final figure
- September Tankan survey and August Indexes of Business Conditions: Whether current business sentiment converges toward the cautious outlook
- Current account from August onward: Whether the widening trade deficit leads to a smaller current account surplus
For a detailed assessment of structural factors, see the quarterly Fiscal Health Check column.
Glossary
| Term | Definition |
|---|---|
| Bear flattening | A flattening of the yield curve during a period of rising yields (a bear market), when short-term rates rise more than long-term rates. It often reflects stronger expectations of policy rate increases. |
| Term spread | The difference between yields on government bonds of different maturities. The 10Y–2Y spread is used as an indicator of the business and policy cycle, while the 30Y–10Y spread is used as an indicator of the fiscal risk premium at the ultra-long end. |
| Real interest rate | The interest rate calculated by subtracting the rate of inflation (year-on-year CPI) from the nominal yield. A negative real rate favors borrowers and indicates financial repression, transferring income from savers to debtors. |
| r−g differential | The difference between the effective interest rate on government debt (r) and nominal GDP growth (g). If negative, the debt-to-GDP ratio is more likely to decline even when the primary balance is in equilibrium. |
| Gross financing needs | The total funding a government must raise from the market over a given period, comprising the fiscal deficit plus maturing debt to be refinanced. It is an indicator of rollover risk. |
| Average remaining maturity | The average time until JGBs mature. The longer the maturity, the longer it takes for changes in market interest rates to pass through to debt-service costs. |
| Interest expense on reserves | The interest the BOJ pays on financial institutions’ current account balances. It increases when the policy rate rises and weighs on the BOJ’s earnings. |
This column was automatically generated by AI integrating Ministry of Finance JGB interest rate data (and tax revenue data when available), Bank of Japan statistics, and e-Stat public statistics as a fiscal analysis resource. This is not a recommendation to buy or sell any financial instruments or government bonds. Please make investment decisions at your own responsibility and consult professionals as needed.