When the bank says no: what a downturn does to the smallest firms' credit, and who answers

Every recession recorded on this site has a chapter about what the banks did: the meetings, the bail-ins, the rescues, the balance sheets. None of those chapters is about the smallest borrower, and the reason is simple. A firm with one employee does not issue bonds, does not appear in the Fed’s stress tests, and does not get rescued. It appears in one place in the official data, a survey question, and what that question has shown for thirty-five years is the subject of this page.

What the survey shows

Every quarter since 1990 the Federal Reserve has asked senior loan officers at the largest banks whether they have tightened or loosened their standards for business loans, and it asks separately about large and middle-market firms and about small ones. Plot the two answers over a cycle and they move together but not equally. In every downturn in the series, 1990, 2001, 2008, 2020 and the tightening of 2023, the net share of banks tightening for small firms rose as fast as for large ones and came down more slowly. Standards for small firms were still net-tightening in 2011, two years after the recession ended, and in 2024, a year after the last bank failures. Big firms get their credit back when the crisis ends. Small firms get it back when the memory of the crisis ends, which is later.

The euro area’s equivalent survey, and the European Central Bank’s own poll of small and medium enterprises, show the same shape, and the reason is not prejudice. A bank prices a loan on the information it has, and it has less on a small firm: no audited accounts, no rating, one customer who might not pay, and an owner whose personal finances are the firm’s. In good times that uncertainty is priced in the rate. In bad times it is priced in the refusal.

The cost is the refusal

This is the point that the rest of this site, being about rates and balance sheets, has never made, and it is the one that matters at the bottom of the market. For a large firm a tighter credit market means a higher spread. For a one-person firm it means the bank’s small-business desk asks for two years of trading history, a tax return and a guarantee, and says it will decide in three weeks. The firm’s need was a supplier who wants paying on Friday for an order whose invoice will be settled in sixty days. Three weeks is not a price. It is a no.

That is why “availability” appears in the ECB’s survey as a separate question from “cost”, and why the answers to it move first. A lender that answers the same day, at a rate that would look expensive on a ten-year comparison, is not selling money at that rate. It is selling the interval between the supplier’s Friday and the customer’s sixtieth day, and the value of that interval to a firm that would otherwise lose the order is not measured in annual percentage points at all. Whether such a loan is a good idea depends entirely on whether the invoice is real. That is the borrower’s question, and no lender answers it.

The case in point

Poland is the clearest example in Europe, because it has more of these firms than anywhere else relative to its size: about two and a half million one-person businesses, sole traders in the British sense, who between them account for a large share of construction, transport, retail and the trades. Polish banks’ small-business desks behave exactly as the surveys predict, and the Polish government’s own reforms of the sole trader’s tax and health-contribution regime since 2022 have made the household side of that balance sheet harder to plan, which tightens the firm side too. Into that gap a regulated non-bank market has grown up in the last decade, supervised since 2024 by the financial regulator, whose products for firms are small, unsecured and decided within a day. Wandoo Biznes is one of the online ones, lending up to twenty thousand zloty to sole traders without collateral, which is to say it sells the interval described above and nothing else. The same shape exists in Britain, Spain and Italy under different names; Poland simply has the most customers for it.

The other side of the ledger

None of this is an argument that the smallest firms should borrow more in a downturn. It is an argument about what a downturn does to them that the aggregate data hides. When the Fed tightens, the newspapers report the rate. The sole trader experiences the refusal, months before the rate would have mattered and months after the rate has come back down. Any account of a recession that stops at the banks’ balance sheets has missed the borrower who never got onto one.

A note on the cycle. The senior loan officer survey for the third quarter of 2026 shows standards for small firms roughly unchanged and demand weak, which is what the late stage of an easing cycle usually looks like. The next tightening will show up in that series before it shows up anywhere else. When it does, the pages in this section will have their next chapter.