The price of money, 2021 to 2026

1. Two prices

A company that needs money has two ways to pay for it. It can sell part of itself, and pay in ownership. Or it can borrow, and pay in interest. Each has a price, and in private markets neither price is published in one place.

This report puts the published pieces together. On the equity side, the best US series comes from Carta, whose cap-table software records the rounds of more than 50,000 US companies.² It is a sample, weighted to venture-backed companies, but it is large and consistent. For the UK we use the British Business Bank's tracker of equity deals in smaller businesses.⁵ On the debt side we use Lincoln International's index of senior loans it values, the filings of the largest listed lenders, and four separate measures of default.

All figures are as their sources publish them. Several sources revise their own history as late deals are recorded. Where that matters, we say so.

2. What equity costs

The cost of equity is dilution: the share of the company the new investors receive. Carta defines it as the round size divided by the pre-money valuation.³

In the first quarter of 2026, the median US round sold about 19% of the company at seed and about 18% at Series A. It sold 13% at Series B, 9.5% at Series C and 4.9% at Series D.¹

The striking thing is the direction. "Median dilution has now fallen three to five percentage points across every stage over the past three years," Carta wrote in its Q1 2026 report.¹ At Series B, the median was about 18% at the end of 2023 and 12.9% at the end of 2025. Across seed to Series C, it fell from about 19% to 16%.²

Founders are giving up less of their companies per round, not more. That runs against the story of a hard market. The explanation is in the round sizes and valuations. The median Series A round was $7.8m at the end of 2022 and $13.5m at the end of 2025. The median Series C post-money valuation reached $385m in early 2026, the highest in Carta's data.^1,4 Investors are paying more for the companies they back.

The catch is the word "they". The averages describe the companies that raised. They say nothing about the ones that did not. Carta's own data shows the time between rounds lengthening: about two years from seed to Series A in 2025, against eighteen months five years earlier.² In late 2023, about 45% of Series A rounds were bridges, extensions of the previous round raised because the next one could not be. That was the highest rate Carta had recorded.³

So equity has got cheaper for the companies that can raise it, and harder to get for the rest. The market has become more selective, not more generous.

3. Down rounds

A down round prices a company below its last round. It is the clearest sign that a company raised at a valuation it could not grow into.

In every quarter of 2023, between 19% and 20% of US venture rounds on Carta were down rounds. The peak came in the first quarter of 2024, at 22.2%. By the first quarter of 2026 the share had fallen to 11.4%, "the lowest figure in nearly four years", and 96.1% of the capital raised that quarter went into up rounds.^1,3

The down-round wave of 2023 and 2024 was the bill for 2021. Companies that raised at the peak met a market that would no longer pay those prices. Most of them have now either grown into their valuations, repriced, or stopped raising. What is left is a market where the median round is an up round and the companies in it are, on average, stronger.

4. The UK

The UK record is thinner but tells a sharper story. The British Business Bank tracks announced equity deals in UK smaller businesses, using Beauhurst's data.

Equity investment peaked at £20.6bn in 2021. It was £10.8bn in 2024, and the number of deals had fallen by a third from the peak, "in line with levels last seen in 2018".⁵ The average pre-money valuation fell further, from £40.2m in 2021 to £20.4m in 2024.

Within that average the stages moved differently. Seed valuations barely changed between 2023 and 2024, at about £5.6m. Growth-stage valuations rose by more than a quarter, from £60.0m to £76.5m. As in the United States, money is concentrating on the companies further along.

Europe as a whole followed the same path. Atomico puts European tech investment at a record $100bn in 2021 and about $44bn in 2025.⁶ It also found that a European seed round is half the size of the US median, and that 51% of European founders said raising had become harder in 2025, down from 60% the year before.

5. What debt costs

The cost of private debt is easier to measure, because it is a rate. The most useful public series is Lincoln International's senior debt index, a fair-value index of the first-lien, unitranche and second-lien loans Lincoln values for its clients.

Its yield rose from 11.1% at the end of September 2022 to a record 11.8% a year later, as base rates rose. It fell to 9.56% by the third quarter of 2025.⁷ More telling is the spread, the part of the yield that is the lender's price for risk rather than the base rate. By the third quarter of 2025 it had fallen to 6.17% over SOFR, "the lowest spread since inception of the LSDI". New unitranche loans to companies with $40m to $100m of earnings priced at about SOFR plus 4.50% at the end of 2025, with an upfront discount of about 1%.⁸

The largest listed lender shows the same path in its own books. Ares Capital's average yield on its debt investments at cost was 8.7% at the end of 2021, 12.5% at the end of 2023 and 10.3% at the end of 2025.¹¹

There is a reason the price fell. Private credit has raised a great deal of money, and lenders compete for the best borrowers. "There's a lot of capital in the market, all chasing high-quality deals," one Lincoln managing director told Fortune in February 2026.²⁰ Our own reading of the loan books of 165 US business development companies finds a median margin of 500 basis points at the end of June 2026. Borrowers with seven or more lenders paid half a point less than borrowers with one.

6. The strain beneath the price

If falling spreads were the whole story, private credit would look comfortable. It is not the whole story.

Average buyout leverage among the companies Lincoln values was 5.2 times earnings at the end of 2025. During 2025, 21% of those companies earned less than their fixed charges.⁸ The share of loans paying some interest in kind rose from 7% in 2021 to 11% in 2025. More revealing is what Lincoln calls "bad PIK": loans where payment in kind was added after closing, usually as a concession to a borrower that could not pay cash. That share rose from 2.5% of loans at the end of 2021 to 6.4% at the end of 2025.⁸

Lincoln calls bad PIK a proxy "shadow default rate". In November 2026 it wrote: "While the 2.5% covenant default rate is encouraging, it tells only part of the story ... The 6% of loans with Bad PIK may provide a better indication of underlying credit stress."⁹

Cheaper debt and rising strain arrived together. The spread fell to its lowest while the share of borrowers unable to pay cash interest rose to its highest. Both are true, and they are connected. Lenders with money to deploy compete on price for new loans, and on patience for old ones.

7. Four meanings of default

The word "default" is used for four different things in US private credit, and the figures are not comparable.

Proskauer's index, the broadest definition, counted 2.41% of loans in default in the third quarter of 2026.¹² KBRA's middle market measure was 2.2% by value in the first quarter.¹⁴ Lincoln's covenant default rate was 2.5% in the third quarter.⁹ Fitch's US private credit default rate reached a record 6.0% in the twelve months to April 2026, according to Bloomberg's report of it.¹³ Fitch's figure is higher mainly because it counts interest deferrals and payment in kind, which made up more than half of its defaults.

Anyone quoting a default rate for private credit should say whose it is. The difference between 2% and 6% is almost entirely a difference of definition.

8. How the terms moved with events

Four events since 2021 show how fast the price of money responds.

The rate rises of 2022 and 2023. The Federal Reserve raised its target range from 0.25% to 5.25-5.50% between March 2022 and July 2023. The Bank of England took Bank Rate from 0.25% to 5.25%.²¹ Private credit yields rose almost point for point. Ares Capital's went from 8.7% to 12.5%. Equity fell harder: Carta's deal value fell by more than half in 2022, and the median Series C round fell from $56m to $21m between the end of 2021 and the end of 2022.⁴

The failure of Silicon Valley Bank, March 2023. "On March 9, SVB lost over $40 billion in deposits," the Federal Reserve's review recorded.¹⁶ SVB held about a fifth of US venture debt by value. Non-bank lenders were expected to fill the gap "at higher rates and tighter terms", and they did.¹⁷ Hercules Capital, the largest listed venture lender, reported a core yield of 11.2% at the end of 2021 and 14.3% at the end of 2023, though part of that rise was base rates.¹⁸

The rate cuts of 2024 and 2025. As rates fell, private credit yields followed them down. Down rounds halved. Default measures rose a little, as borrowers who had held on through high rates ran out of room.

The tariffs of April 2025. After 2 April, lenders asked for an extra 0 to 50 basis points, more for smaller and tariff-exposed companies, and unitranche loans priced at SOFR plus 4.75% to 5.75%.¹⁰ By the third quarter of 2025 the shock had passed and spreads were at record lows.⁷ The market priced the uncertainty in days and forgot it in months.

Rates are now rising again. The Federal Reserve raised its range to 3.75-4.00% in September 2026, and the Bank of England took Bank Rate to 4.00% on 5 November 2026.²¹ It is too early for the terms data to show the effect.

9. Money for the lenders

One more change bears on everyone who borrows privately. The funds that lend are themselves finding it harder to raise. Credit funds that closed in the first half of 2026 took an average of 39 months to reach a final close, against 26 months in 2024. The ten largest took 64% of the capital raised.¹⁹

Private credit is concentrating in fewer, larger managers. For borrowers, that means more of the market's price is set by a handful of lenders who see most of the deals. It also means smaller managers, with smaller funds, are hungrier for the loans the large ones pass over.

10. What it means

For a company raising equity: the price is better than it was, if you can raise at all. The median founder sells less of the company per round than three years ago. The market is selective, the time between rounds is longer, and bridges are common. Plan for the round to take longer than the last one.

For a company raising debt: the margin is the lowest in years, and the terms deserve more attention than the margin. Payment in kind, covenants and call protection decide what a loan costs when things go wrong, and lenders are now writing a great deal of the first.

For anyone reading the numbers: check the definition. Dilution, down rounds, yields and defaults each mean something specific to the source that publishes them. Several sources revise their history. Quote the latest edition, with its date.

Cite as: “The price of money, 2021 to 2026”, Norvian Review, 6 November 2026.