The private credit expansion is one of the most consequential structural stories in finance over the last decade, and the answer to whether it is durable has gotten more complicated in just the last few months.
Private credit assets under management have grown from roughly $158 billion in 2010 to somewhere between $1.7 and $2.3 trillion today, with Preqin projecting it will nearly double to approximately $4.5 trillion by 2030. Direct lending — corporate loans to mid-sized, usually private-equity-backed companies — is the core of it, but it has now grown large enough that direct lending roughly matches the broadly syndicated loan market in size, at $1.5 to $2 trillion. The largest managers — Blackstone, Apollo, KKR, Ares, Blue Owl — have made credit their largest and fastest-growing business line. Apollo’s credit book alone reached $723 billion and now comprises approximately 82% of the firm’s total AUM.
The expansion was not an accident of risk appetite. It was a mechanically predictable response to a regulatory capital constraint.
The Regulatory Arbitrage
Post-2008 Basel rules made leveraged and acquisition finance progressively more expensive to hold on a bank’s balance sheet. The 2023 Basel III “Endgame” proposal would have pushed that further, with estimated aggregate capital increases of roughly 16% for the largest U.S. banks and as high as 21% for the globally systemic ones. Non-bank lenders do not carry that regulatory capital burden, so they can hold the same loan more cheaply.
That gap is the single biggest reason private credit could out-price and out-compete banks in the middle market over the last decade. The “banks are losing the middle market” narrative is true at the smaller end: direct lenders now finance roughly 90% of buyout transactions below $500 million in enterprise value, and this happened before any final rule was even in force. Credit committees adjusted behavior in anticipation of the regulation, not in response to it.
At the large-cap and upper-middle-market end, the picture is more nuanced. Pricing has converged as both camps chase the same deals — global new-issue direct loan spreads compressed from approximately 716 basis points in early 2023 to roughly 544 basis points by year-end 2025. Banks have been actively reclaiming share here: syndicated spreads for large-cap borrowers compressed to roughly SOFR+300, materially tighter than the roughly SOFR+480 and SOFR+500 that large-cap and middle-market direct loans were pricing at. Some of the largest recent LBO financings have gone to bank syndicates specifically because banks won the price war.
The Regulatory Ground Shift
The regulatory constraint story is no longer a static fact. On March 19, 2026, U.S. federal banking agencies reversed course: they issued a package of proposals that would formally rescind the 2023 Basel III Endgame proposals and instead cut CET1 capital requirements by up to approximately 7.8%, explicitly because the old framework had become too duplicative and punitive, pushing activities like middle-market lending toward nonbanks.
That reproposal is still in comment period and is not final, but directionally it makes it cheaper for banks to hold exactly the assets private credit had been winning. The next two to three years of market share numbers will be a real test of how much of private credit’s edge was structural versus regulatory-arbitrage.
Re-Intermediation, Not Retreat
Banks are not simply retreating from the middle market. They are re-architecting how they participate. Rather than a binary “banks versus non-banks” fight, what has actually happened is a re-intermediation: banks increasingly originate, structure, and provide fund-level financing — subscription lines, warehouse facilities, NAV lending — to the private credit funds that hold the actual loans, which is far more capital-efficient than holding the loan directly.
Wells Fargo partnered with Centerbridge in 2023, followed by Lloyds–Oaktree, Citi–Apollo ($25 billion over five years), PNC–TCW, and Webster–Marathon. Apollo has also brought JPMorgan, Goldman Sachs, and other banks in as broker-dealers to trade and syndicate its investment-grade private loans. Banks keep the origination relationship and fee income, private credit managers get deal flow and balance-sheet scale, and everyone avoids holding the riskiest tranche on a regulated balance sheet.
The Liquidity Stress Test
The industry spent years marketing itself as immune to the liquidity mismatches that hit public credit — long lockups, loan maturities matched to fund life. That claim got tested in early 2026.
Blue Owl restricted redemptions on one of its BDCs in February 2026, converting quarterly tender offers to a locked return-of-capital structure after facing double-digit percentage redemption requests. The announcement triggered a selloff that spread to Apollo, Blackstone, KKR, and Ares, and reignited debate over liquidity, valuation transparency, and systemic risk. Several other large evergreen vehicles — Blackstone’s BCRED, Apollo’s ADS, Cliffwater — have also seen elevated redemption requests and slower fundraising since, and managers have had to sell loan portfolios or inject their own capital to meet redemptions rather than gate outright.
Regulators are watching this closely. The Financial Stability Board’s May 2026 report flagged four vulnerability clusters: bank interconnections, borrower credit quality and valuation opacity, concentration and liquidity mismatches, and data gaps. It separately noted that private-credit CLOs, now roughly $155 billion outstanding, tend to carry a higher share of CCC-rated loans, lower weighted average ratings, and weaker recovery rates than traditional balance-sheet CLOs. Fed Vice Chair Bowman told Congress in June that it has been very difficult for regulators to get a clear picture of where bank funding into the non-bank credit space is actually flowing.
Where the Edge Is
The structural shift is real and mostly capital-driven, not a temporary cyclical story. Banks’ return-on-capital economics for holding middle-market credit were genuinely worse than a fund’s, and that gap opened up over more than a decade of tightening bank regulation. But the regulatory story has flipped: the same capital rules that pushed lending out of banks are now, in proposal form, being loosened specifically to pull some of it back.
The growth has not been risk-free. Rising defaults, the Blue Owl episode, and FSB/Fed concern about interconnection and opacity are the first real stress test since most of this AUM was raised. Fed and industry voices so far are converging on a painful sector-specific shakeout rather than a systemic crisis, though that assessment could change if redemption pressure keeps building.
The alpha is not in the generic constraint — it is in the specific layers that enable the constraint to be resolved. In this case, it is the fund-level financing infrastructure (subscription lines, warehouse facilities, NAV lending), the origination and structuring relationships between banks and private credit managers, and the valuation transparency layer that regulators are demanding. The re-intermediation is the real story, not the retreat.