Howard Marks: Private credit's 1929 echo
Howard Marks' April 2026 Oaktree memo traces private credit's $2 trillion expansion through the innovator/imitator/idiot arc, surfaces three structural echoes of the 1929 crash, and offers a conviction-vs-imitation calibration framework for direct lending investors.
Published: 2026-04-09
Warren Buffett's version of the thought comes in three words. "First the innovator, then the imitator, then the idiot." 1
Howard Marks (co-founder and co-chairman of Oaktree Capital Management, one of the world's largest distressed-debt investors) quotes that line near the top of his April 9, 2026 memo What's Going on in Private Credit?, and it does the work the rest of the piece unpacks: private credit's fifteen-year boom followed the script precisely, and the idiot phase is now arriving. 1
From $150 billion to $2 trillion in fifteen years
Marks traces the credit market through seven distinct phases: the 1970s acceptance of non-investment-grade debt, the 1980s leveraged-buyout wave, the 1990s rise of syndicated lending and securitization, the 2000s alternative-investment craze (and the subprime disaster it produced), and then, in the 2010s, the rapid expansion of direct lending. The final stage arrived in the 2020s: direct lending vehicles were packaged and sold to individual investors and retirement accounts. 1
The numbers are stark. The entire private credit industry stood at roughly $150 billion twenty years ago. In the fifteen years since, approximately $2 trillion in direct loans has been made. 1 The mechanism was familiar: early direct lenders demanded high rates and strong protections because banks had stepped back and demand for private financing was high. Institutional investors noticed the attractive early loans and joined. Competition arrived, driving down yields and loosening loan terms. The 17 years of benign conditions after the Global Financial Crisis let the loosening go unnoticed.
One number in the memo stands out as a signal of how much the category drifted from its original credit profile. Software debt now accounts for 20–30% of direct lending portfolios, compared to just 4–5% of the high-yield bond market. 1 That concentration is partly what made the February 2026 AI-driven selloff in software valuations so disruptive for direct lending funds — and partly why it is only a preview of what a full credit cycle would look like.
Marks is careful to add a constraint that applies to all negative readings of private credit sentiment right now:
"Investors are rarely the informed, methodical, dispassionate weighing machine Benjamin Graham and David Dodd described in Security Analysis... in real life things fluctuate between pretty good and not so hot, but in the minds of investors they go from flawless to hopeless." 1
The pendulum swings both ways. Direct lending is not going to zero. The question is how deep a correction is needed to reprice risk honestly.
The three echoes of 1929
Marks draws a structural comparison to the 1929 crash, citing Andrew Ross Sorkin's book 1929: Inside the Greatest Crash in Wall Street History — and How It Shattered a Nation. 1 The 1929 collapse had three enabling features; all three are present in today's direct lending market.
First, unsuitable investment products were sold to the general public without adequate disclosure of the risks involved. Today's direct lending vehicles — packaged for individual and retirement investors — share that trait. Individual investors rarely have the financial sophistication to model the credit risk embedded in a portfolio of middle-market corporate loans, and the marketing of these products typically led with yield, not with the structural complexity underneath it.
Second, buyers were offered substantial leverage to amplify returns. Leverage magnifies losses as readily as gains; in Marks' telling, that fact was consistently underemphasized during the distribution push.
Third, and most structurally dangerous, there was an illiquidity mismatch: the underlying assets cannot be easily sold, but the financing used to buy into these vehicles carried shorter time horizons. When sentiment turns, investors discover that exit is far harder than entry. Sorkin's summary of the enduring lesson applies directly:
"The enduring lesson is not that booms can be prevented, or that busts can be fully averted. It is that we need to remember how easily we forget. The antidote to irrational exuberance is not regulation by itself, nor skepticism, but humility — the humility to know that no system is foolproof, no market fully rational, and no generation exempt." 1
Marks adds a more operational image with the Chuck Prince quote. The former Citibank CEO, speaking in 2007 just before the financial crisis, put it plainly: "As long as the music is playing, you've got to get up and dance." 1 For fund managers raising capital, the logic is not irrational: sitting out the boom means losing allocations to competitors. But the dance floor fills with people who arrived late, paid high, and agreed to unfavorable terms. That is the mechanism by which "the wise man does in the beginning" becomes something the fool does at the end.
Where this leaves the reader
Marks offers a framing for how to hold uncertainty without either full conviction or full avoidance:
"True believers make the most money in manias, and skeptics lose the least when they crash. But the key to the investment success we aim for lies in always maintaining a healthy balance between belief and doubt." 1
The practical calibration Oaktree itself has used: direct lending accounts for less than half of Oaktree's private credit allocation, and private credit accounts for less than half of its performing credit overall — meaning direct lending sits at roughly 15% of Oaktree's total assets under management. 1 That is not a model to copy mechanically, but it illustrates where the firm that wrote the memo has placed its own money.
For readers with private credit or direct lending exposure in their portfolios, Marks' memo suggests one clarifying question worth asking: does the size of that allocation reflect an active conviction about yield and credit quality — or did it grow because peers were allocating and the early returns looked compelling? The distinction matters most not when the music is loud, but when the dance floor starts to thin.
Marks closes with the core mechanism that governs every credit cycle: "In my experience, the limiting factor in the credit markets is never borrowers' appetite for capital, but rather lenders' willingness to supply it." 1 When that willingness snaps back, pricing resets — and that is when the allocation decisions made during the boom become consequential.
The full memo, including podcast audio (39 minutes), is available at oaktreecapital.com.
References
- 1What's Going on in Private Credit? — Howard Marks
oaktreecapital.com
This story was produced automatically by a channel. One sentence is all it takes for Neodrop to keep producing for you.