Minsky Moments in Venture Capital (summary)
Summary of an essay by Abraham Thomas, published in Pivotal on 12 February 2022. Read the full essay on Pivotal →
Can Minsky cycles happen in venture capital? Abraham Thomas argued in February 2022 that they can, and that in venture the variable driving the cycle is time. In Minsky’s framework, capital flowing into an asset class lowers its perceived risk, which attracts more capital. In venture, faster markups, faster rounds and faster fund deployment had made the venture J-curve disappear, so the asset class looked far less risky than before. If funding cycles slowed, the loop could run in reverse.
The essay was published at almost exactly the peak of the 2021–22 startup funding bubble. Three months later, the Financial Times’ FT Alphaville (“A Minsky moment for venture capital?”, Robin Wigglesworth, 4 May 2022, pdf) reported that Refinitiv’s venture capital index had fallen 24.2% in April 2022 and 45.8% year to date, its worst month since the dot-com bust. It quoted Thomas’s conclusion at length, and called it “pretty on the money”. In 2025 Thomas wrote that his timing had been “spot on” and that his predictions about lengthening timescales “proved to be very accurate”.
What is a Minsky cycle?
Hyman Minsky’s “financial instability hypothesis” explains boom-and-bust cycles in public markets. Thomas’s summary: the key idea is not that rising prices attract capital, but that capital inflows reduce perceived risk.
He illustrates it with his own career as a bond arbitrageur at a quant hedge fund:
- Alpha decays. Spreads of tens or hundreds of basis points when he started trading shrank to single basis points within a decade.
- Institutions add exposure as returns fall. Faced with a tenfold drop in expected value, many portfolio managers increased their exposure tenfold to keep their dollar P&L, a rational response to their incentives.
- Risk models encourage it. Arbitrage capital damped volatility, so bond arbitrage looked much less risky in 2006 than in 1999: returns fell, measured risk fell faster, and Sharpe ratios rose. That is a Minsky boom.
- Then the reversal. A shock slows the inflow, prices soften, and three spirals follow: a risk spiral (positions trimmed as volatility rises), a margin spiral (leveraged investors forced to sell), and a redemption spiral (investors pulling money out). Credit markets did exactly this in 2007–08.
“Stability breeds instability, and vice versa.”
Why might venture capital seem immune?
Venture funds use no leverage, offer no redemptions, face no margin calls or counterparties, and benefit from volatility. So there are no risk, margin, or redemption spirals. What, then, could drive a Minsky boom or bust?
Why is time the key variable in venture?
In 2021–22, every part of venture was speeding up:
- startups were being marked up faster than ever;
- funding rounds were closing faster than ever;
- funds were being deployed faster than ever (one LP cited 33% of capital called in a fund’s first year in 2021).
Accelerated markups made the venture J-curve disappear. Thomas’s worked example:
| Scenario | After 18 months, of 10 startups | Portfolio value vs. cost |
|---|---|---|
| Earlier era | 3–4 raise one further round, each at a 2x markup; the rest fail or are doomed | 0.6–0.8x (the dip of the J-curve) |
| 2021-era | 3–4 survivors each raise 2–3 more rounds at 2x markups | 1.2–2.4x |
Fast markups hid the losers, raising both the upper and the lower bound of returns, so venture looked almost riskless. Higher returns and lower apparent risk drew in new money, which sped markups further: “the classic template for a Minsky boom, and it’s all driven by compressed time.” Fast markups also inflate interim IRR, which firms use to raise new funds faster.
What is the difference between measured risk and true risk?
Minsky cycles are driven by the gap between the two. A lender’s true risk is default, measured by proxy through credit spreads, and spreads also reflect demand for credit. A venture investor’s true risk is startup failure, measured by proxy through markups, and markups also reflect demand for startup equity. Booms lower measured risk; busts begin when the market realizes true risk hasn’t gone away.
The deciding question, as Thomas framed it: “Does the compression of timelines in venture change the distribution of terminal outcomes for venture-backed companies?” If it does, the boom was a rational maturing of the industry. If it doesn’t, venture was in a Minsky boom awaiting a Minsky bust.
What could trigger a venture Minsky bust?
Venture already has a known death spiral: the down round. It causes a valuation spiral (early investors and common shareholders wiped out, underwater options), usually alongside a talent spiral (good people leave, weaker hires replace them). To avoid down rounds, startups wait: they cut costs and try to “grow into their valuation”.
Thomas argued that this waiting could be the trigger. If compressed timelines drive the inflows, then anything that delays funding cycles reverses them. Startups delay raising; VCs then lack the markups to raise their own next funds; LPs, seeing weaker recent returns, reconsider their allocations; and capital leaves the asset class. “Minsky giveth, and Minsky taketh away.”
What idea did this essay start?
A footnote proposed that “if Wall Street is in the business of spatial arbitrage, Silicon Valley is in the business of temporal arbitrage.” Thomas developed it into a full theory of venture capital in Making Markets in Time (2025).
Related
- Full essay: Minsky Moments in Venture Capital, Pivotal, 12 February 2022
- Making Markets in Time (summary)
- The Perils of Prudence (summary)