Labour market
Small business hiring intentions fall to a 2016 low as credit does the tightening
Net hiring intentions on the Marlowe Institute small business survey dropped to 8 per cent, and the firms reporting harder credit conditions are the same ones planning to shrink headcount. The transmission from the long end of the curve runs through here.
In this story
Illustrative. Bank Season is an editorial prototype. This story, its sources, the issuers named in it and every figure it quotes are invented to demonstrate the publication. Nothing here is reported fact or investment advice. Read the disclosure.

Large firms borrow in the bond market, where a single-A issuer paid a four basis point concession to raise money this week. Small firms borrow from a bank, at a spread over a floating benchmark, on a facility that gets reviewed annually. Those two funding markets have moved in opposite directions since June, and the Marlowe Institute survey of 4,200 firms with fewer than 500 employees is the clearest measure of what the second one is doing to employment.
8%
Net hiring intentions, Marlowe small business survey
August, from 19 per cent in February. The 2016 average was 9 per cent.
Net intentions means the share of firms planning to add staff over the next three months minus the share planning to cut. Eight per cent is a positive number and a firm labour market can live with it for a while. What makes it uncomfortable is the speed of the descent and the fact that the answer is almost entirely explained by one other question on the same survey.
The credit question explains the hiring question
Marlowe asks respondents whether credit is easier or harder to obtain than three months ago. The share answering harder rose to 34 per cent in August, from 21 per cent in February and the highest since the regional bank episode of 2023. Split the hiring intentions by that answer and the aggregate number stops being a single figure and becomes two very different ones.
Marlowe small business survey, August responses by credit condition
| Reported credit condition | Share of firms | Net hiring intentions | Planning capital spending | Reporting margin compression |
|---|---|---|---|---|
| Easier than three months ago | 9% | +31% | 44% | 18% |
| Unchanged | 57% | +16% | 29% | 34% |
| Harder than three months ago | 34% | −9% | 11% | 71% |
| All respondents | 100% | +8% | 25% | 44% |
Survey of 4,200 firms with fewer than 500 employees, conducted August 4 to August 22.
A third of the sample is planning net job cuts. That group is also the group reporting margin compression at twice the rate of everyone else and planning capital spending at a quarter of the rate. These are firms adjusting to a higher cost of capital by shrinking, which is the textbook transmission mechanism working exactly as designed. The complication is that the policy rate has fallen 75 basis points this year, so the cost of capital they are responding to is not the one policy sets.
Net hiring intentions and firms reporting harder credit
Per cent, Marlowe Institute small business survey, monthly.
The two lines cross in April and keep separating. Small business borrowing is priced off a bank's cost of funds and its capital charge, neither of which improves when the front end of the Treasury curve rallies. What has happened instead is that the ten-year yield has risen 62 basis points this year, banks have repriced term lending against it, and covenant tests written against 2021 cash flows have started to bite on firms whose revenue has not grown.
The committee cut the policy rate three times and small business credit got tighter every quarter. That is not a policy failure, it is a reminder that the rate small firms borrow at is set at the ten-year point and always has been.
Dov Ackerman, chief economist at Bellweather Macro
The panel data agrees
Survey answers describe intentions. The Verity Research payroll panel, which processes payroll for about 14 per cent of US private employment, describes what firms actually did. It has private job growth averaging 31,000 a month across the summer against 148,000 in the same three months of 2025. Split by employer size, firms with fewer than 100 employees contributed nothing to that average, and firms between 100 and 500 contributed 4,200 a month.
31,000
Average monthly private job growth, June to August
Verity Research panel. Firms under 100 employees contributed none of it.
Health care, utilities and defence adjacent contracting supplied 27,400 of the monthly average, and all three are dominated by large employers whose demand comes from budgets and demographics rather than from the credit cycle. Strip them out and the private economy is adding a few thousand jobs a month, almost all of it at firms large enough to fund themselves without a bank.
- The quits rate on the Verity panel has fallen to 1.8 per cent, a level last seen in 2016, and it falls fastest among workers at small employers who have fewer outside offers.
- Temporary help employment has declined for six consecutive months, and it is the only component of the panel that reliably leads the aggregate.
- Initial claims remain near 224,000 on a four week average, confirming that firms are freezing hiring rather than cutting staff.
- The Russell 2000 has underperformed the S&P 500 by 9.4 per cent since July, which is the equity market pricing the same funding condition the survey is measuring.
- Firms reporting harder credit are concentrated in construction, retail trade and professional services, which is where the panel already shows outright job losses.
Nothing in this survey says recession. It says the marginal employer has stopped, and the aggregate is being held up by three sectors that do not respond to interest rates at all.
Marguerite Paz, senior economist at the Marlowe Institute
Policymakers meet on Wednesday and are widely expected to leave the policy rate alone while addressing the balance sheet. If the reinvestment decision pushes long yields higher rather than lower, the transmission described here gets worse rather than better, and it does so through a channel the statement will not mention. The survey field period for the September round opens on October 5, which is the first reading that will contain the answer.
The honest caveat is that small business surveys have signalled trouble that did not arrive on at least three occasions since 2011, and sentiment measures carry a persistent political tilt that makes level comparisons across administrations unreliable. The credit question does not have that problem. Whether a bank renewed a facility is a fact rather than an opinion, and a third of the sample is reporting that it got harder.


