As interest rates rise, firms buckle. Put July’s run of regional-bank losses beside the collapse of Zentoshin, an Osaka payment agent, and the story writes itself. It is half wrong. The Bank of Japan did raise its policy rate to 1.00% in June, the highest since 1995, but stripped of inflation the real rate is still deeply negative (Dai-ichi Life Research Institute). A rate of 1% is not, in itself, crushing borrowers. What changed is not the level of rates but the machinery that ran beneath them.
The numbers are indeed worsening. Corporate failures in the first half of 2026 came to 5,335 by TDB’s count and 5,346 by TSR’s, up 6 to 7% on the year and above 5,000 for a second straight year on TDB’s count and the first time in 12 years on TSR’s (TDB, TSR). But look inside and it is not a story about rates. Small failures, under ¥50m of debt, were 62% of the total, and weak demand was the leading cause in 80% of cases. Failures attributed to higher costs (439), to labour shortages (237) and to unpaid taxes (126) each set a record for their series. Meanwhile failures among firms that took the pandemic’s zero-zero loans, the interest-free emergency credit of 2020-21, fell 26%: the effect of that particular support has run its course. What is breaking is cost, labour and demand, not interest. The cast of the cull is shifting, from the after-effects of the pandemic to structural pressure.
Why now? In the era of zero rates and ample liquidity three things held at once: weak borrowers could roll their debt and survive; yield-starved lenders could lend loosely; and unfilled holes could be carried forward as long as new money kept arriving. Normalisation takes those props away one by one. Higher rates narrow the room to roll; as liquidity drains, loose underwriting and carried-forward holes stop working; and the subcontracting-fairness law that took effect in January 2026 closed the route by which large firms pushed their financing strain onto suppliers through notes and factoring (JFTC). The rate is not pulling the trigger. It is removing, plank by plank, the scaffolding that held borrowers up. The same rising costs and labour shortages turn into failures once the support is gone.
There is an objection, of course. The leading causes are cost and labour, nothing to do with normalisation; talk of regime change is a story told after the fact. Half of that is right. The trigger is cost and labour. But the same cost pressure, a few years ago, with rollover and loose credit and inflow all working, would not have converted into failure on this scale. What changed is not the force of the pressure but the thickness of the cushion beneath it.
“The Cull You Can’t Count” traced one of those props last month. As banks and promissory notes withdrew, small firms’ working capital moved into receivables factoring and revenue-based finance, a corner with no statute of its own and one that barely registers in the statistics. What that layer has in common is that it works only while money keeps flowing: turn a receivable into cash now, cover it with the next receivable. While the flow is fast, the hole stays out of sight.
Zentoshin was the extreme case. What it did was, in function, factoring of card receivables: it advanced merchants their card takings and raised the working capital for it from regional banks with no borrowers at home. But Zentoshin is not a clean example. On TSR’s account it had falsified its books for at least 20 years and was insolvent by some ¥60.5bn. So one cannot say the rate killed it: what killed it was a hole of its own and the funding that dried up after the 2024 criminal case. Even so it concentrates the pattern in one place: credit that lives on flow, regional banks starved for yield and a trigger that stops the flow. The trigger was not the rate; it differs from firm to firm.
So it is early to treat any of this as proof that rate rises are breaking Japan. The policy rate is 1%, the terminal rate that markets and brokers see is around 1.5% (Nikkei/QUICK survey), and the real rate is still negative. What is surfacing now is the layer most dependent on the old regime and most deeply hidden. The level of rates is not, broadly, biting. At the entrance to normalisation, what was overstretched is simply showing itself first, in order. This is exposure firm by firm, intermittent, not a cascade.
For investors the implication sits with the regional banks. Normalisation lifts their margins; that is sound, and it is the bull case. But the same normalisation is bringing to the surface, one by one, the risky out-of-area, out-of-sector credit that earned those banks a yield beyond their margins. The improving margin and the deteriorating credit are two sides of one coin. The provisions taken by the six banks that lent to Zentoshin, Towa’s ¥5.886bn among them, are only the first measured instances of the far side. The bulls are looking at one face of the coin. For where the next one surfaces, watch the non-bank advance-and-factoring layer and the thinly disclosed out-of-area credit of the regional banks. Zentoshin’s creditor list already carried an example of the first: a revenue-based crowdfunding lender, at ¥2.1bn, with retail money behind it (Nikkei). There is one more reason not to turn bullish on the regional-bank sector.
(This piece is not investment advice. Zentoshin’s alleged accounting fraud and insolvency are TSR’s findings at this stage and not established fact. Figures draw on published material from the Bank of Japan, TDB and TSR and on the banks’ own disclosures.)