Comments

Rising interest rates and sustained inflation in Japan will have a limited effect on underlying loans.   Currently, delinquency and replacement/withdrawal rates are on a slight upward trend. Even so, the rates remain quite low. The overall Japan Housing Finance Agency (JHF) series pool's (see note 1) performance also remains relatively stable.

As of May 2025, S&P Global Ratings forecasts inflation of around 2%-3% through 2028. With wage hikes of over 5% on average again expected after spring negotiations this year, we expect inflation to have a limited impact on borrowers' ability to repay debt. The policy rate in Japan has risen to 0.5%. We forecast the rate will rise to 1.5% by 2027. However, a rise in interest rates will not directly affect the performance of underlying JHF loans, which are all fixed-rate mortgages.

We are looking closer at trends in delinquency rates of loans backing monthly notes.  The one-month to four-month delinquency rates of loans backing monthly notes have been rising slowly since 2022. They are now slightly above pre-pandemic levels (see chart 11). These rates for monthly notes issued in and after fiscal 2020 (ended March 31, 2021) are increasing slightly faster than before (see chart 12).

We believe this acceleration of delinquency rates is partly due to changes in borrower attributes. We have seen an increase in self-employed borrowers, rather than civil servants or regular employees. More borrowers now tend to have debt to income (DTI) ratios of 30%-35%, which are higher than before. In addition, pool sizes by fiscal year are declining due to smaller issuance amounts, leading to some fluctuations in delinquency rates. We will closely monitor delinquency rates, which are a leading indicator of replacement and withdrawal ratios.

Replacement and withdrawal ratios for monthly notes are another key focus.  The current rate of about 0.6% is about 0.2 percentage points higher than the pre-pandemic level (see chart 3). As with delinquency rates, replacement and withdrawal rates for monthly notes issued in and after fiscal 2020 have increased at a faster pace. We attribute this partly to changes in borrower attributes. Smaller issuance amounts and pool sizes by fiscal year are behind some of the rate volatility.

Increased economic stress could have a larger impact on performance. However, as of May 2025, we assume that Japan's unemployment rate will remain low and stable at around 2.5% through 2028. This will underpin the performance of mortgage loans. It is therefore unlikely that replacement/withdrawal rates will rise sharply, in our view.

Delinquency and replacement/withdrawal rates for series T notes will remain stable.   Underlying pools backing series T notes have longer seasoning, and redemption of principal payments is further on for these notes than for the monthly notes. Both delinquency and replacement/withdrawal rates remained stable even during the pandemic. They display less sensitivity to stress environments compared with the monthly notes (see chart 6 and 14).

Prepayment rates will remain low amid rising interest rates.  Prepayment rates of loans underlying the monthly notes issued in and after fiscal 2016 have been low, at around 2% for the last few years. Interest rates applied to loans executed since the Bank of Japan began its negative interest rate policy are generally low. In our view, lower incentives to refinance in the face of rising interest rates have eroded prepayment rates (see charts 17 and 18). Meanwhile, prepayment rates of loans underlying series T notes have recently been stable at around 5%-6%. In our view, most outstanding obligors in the underlying pools are less sensitive to movements in interest rates because of the completion of the step-up timing in interest rates of the underlying pools (see charts 20 and 21).

We expect the monthly notes issued in and after fiscal 2016 to have a sizable influence on the current pool's performance.  The outstanding balance of overall monthly notes as of Jan. 31, 2025, is about ¥17.1 trillion. Of this, series issued in and after fiscal 2016 accounted for about 76%, while series issued in and after 2020 amounted to about 36%. However, the outstanding amount has been declining as recent issuance amounts have been shrinking. No recent new issuance in series S and T notes backing loans directly extended by Government Housing Loan Corp. (GHLC), the predecessor of JHF, has led to the decline in the outstanding balance of these notes. It peaked at about ¥4.4 trillion in 2009, and is now below ¥0.2 trillion. (see charts 1 and 2).

JHF monthly notes were fully redeemed up to series 17 notes.   JHF exercised call options on notes for which the outstanding balance had fallen to less than 10% of their initial balance. It exercised options on JHF series 10, 11, 14, 15, 17 notes in April 2025. As of now, JHF series 1 to 17 and all S notes have been fully redeemed.

Pool Composition By Fiscal Year Of Issuance Or Contractual Fiscal Year

Table 1 (a)

Table 1 (b)

Table 2(a)

Outstanding Balance Of Receivables

Chart 1

Chart 2

Performance

Replacement and withdrawal ratios

Chart 3

Chart 4

Chart 5

Chart 6

Chart 7

Chart 8

Cumulative replacement and withdrawal ratios

Chart 9

Chart 10

Delinquency rates

Chart 11

Chart 12

Chart 13

Chart 14

Chart 15

Chart 16

Prepayment rates

Chart 17

Chart 18

Chart 19

Chart 20

Chart 21

Chart 22

Appendix 1: Charts 3 To 22

• Figures for loan pools by fiscal year of issuance or contractual fiscal year are weighted averages of the outstanding balances.
• Breakdown of loan pools by fiscal year of issuance or contractual fiscal year: Our 12-month moving average starts at the first month, which we define as the month after the one in which JHF issued its first series for that fiscal year.
• The rightmost portions of trendlines in some charts with seasoning show volatility until a full 12 months of data are accumulated for each fiscal year.

Appendix 2: Analysis

We detail performance trends in the underlying pools of loan receivables backing the notes that JHF issued up to the end of January 2025 (see note 3). We separately examine:

• The performance of the overall series pool, which consists of all pools of loan receivables backing all regular monthly notes and is mainly made up of loans JHF purchased from private sector financial institutions;
• The performance of the receivables pools backing each individual series of regular monthly notes, broken down by fiscal year of issuance (see note 4);
• The performance of the overall series S and series T pools, which consist of all pools of loan receivables that back all series S and series T notes and were extended by JHF itself; and
• The performance of the receivables pools backing each series S and series T note, broken down by the fiscal year in which the loan agreements were made (see note 5).

1. In this report, "monthly notes" refers to notes that JHF issued unless otherwise noted. The performance data in this report includes that of the former GHLC notes.

2. JHF has a replacement scheme for all GHLC notes and for JHF series S notes. For JHF regular monthly notes and JHF series T notes, JHF has a withdrawal scheme that applies an amount equivalent to the principal on loans close to default and other loans to payment of the notes and withdraws such loans from the entrusted assets. JHF defines receivables close to default and other receivables as:

• Receivables that are four months overdue,
• Receivables that JHF deems defaulted,
• Receivables with extraordinary amendments to terms and conditions,
• Receivables subject to changes in obligor (contracting party) names, and
• Other receivables.

The fiscal 2016 loan pool includes the series 108 monthly notes, for which an administrative error at a partner financial institution led to multiple withdrawals. We thus excluded all "other receivables" for the one month following entrustment regarding the series 108 monthly notes.

3. This report updates "JHF RMBS Performance Watch October 2024: Delinquency Rise Will Be Negligible," published Oct. 29, 2024. We periodically publish surveillance reports on the structured notes that JHF issued. In this report, we have added performance data for the collection period from August 2024 to January 2025.

4. We aggregated the loans underlying the series of regular monthly notes by fiscal year (ends March 31 of the following year) of issuance. As a result of the relatively short ages of the underlying loans at the time of their entrustment, aggregated data by fiscal year of issuance are almost identical to aggregated data by contractual fiscal year (the fiscal year in which the loan agreement was entered into).

5. For series S and series T notes, there was a significant period between extension of the underlying loans and JHF securitizing them. We therefore aggregated the loans by contractual fiscal year.

Appendix 3: Calculation Of Indices

Replacement or withdrawal ratio (annualized)

Receivables replacement or withdrawal ratio for a term "t" = amount of receivables subject to replacement or withdrawal for term "t" (principal)/receivables outstanding at beginning of term "t" (principal) x 100 x 12

Cumulative replacement or withdrawal ratio

Cumulative replacement or withdrawal ratio for term "t" = cumulative amount of receivables subject to replacement or withdrawal from note issuance to end of term "t" (principal)/initial receivables outstanding (principal) x 100

Delinquency rate

Delinquency rate for term "t" = amount of delinquent receivables for term "t" (principal)/receivables outstanding at beginning of term "t" (principal) x 100

Prepayment rate (annualized)

Prepayment rate for term "t" = amount of prepaid receivables for term "t" (principal)/receivables outstanding at beginning of term "t" (principal) x 100 x 12

Related Criteria

• Global Methodology And Assumptions: Assessing Pools Of Residential Loans--Asia-Pacific Supplement, April 4, 2024
• Global Methodology And Assumptions: Assessing Pools Of Residential Loans, Jan. 25, 2019

Related Research

• JHF RMBS Performance Watch October 2024: Delinquency Rise Will Be Negligible, Oct. 29, 2024