E80: Recession deep dive: VC psychology, macro risks, Tiger Global, predictions and more

Multiple national home-price indices show that U.S. residential prices peaked in mid‑2022 and then declined for several months, well within the “next few quarters” after the May 2022 podcast:

• The S&P CoreLogic Case‑Shiller National Index peaked in June 2022 and then fell for at least seven consecutive months; by November 2022 it was down about 3.6% from the June peak, and by January 2023 prices had fallen roughly 5% from that peak. (press.spglobal.com)
• An analysis of Case‑Shiller data in Fortune notes that national home prices peaked in June 2022 and then declined about 3% (seasonally adjusted) and 5% (not seasonally adjusted) over the following months, describing it as a post‑WWII home‑price correction. (fortune.com)
• National Association of Realtors data reported by Axios show that the median existing-home price fell about 6% between June and August 2022, meaning price declines were visible in the standard housing data just a couple of months after the May conversation. (axios.com)
• Over the same period, 30‑year mortgage rates roughly doubled: Freddie Mac’s survey shows the average 30‑year fixed rate rising from around 3% in 2021 to 5.8% in June 2022 and then above 6%–6.7% by September 2022, with Freddie Mac explicitly noting that these higher rates were putting downward pressure on home prices. (freddiemac.gcs-web.com)

These data confirm that U.S. home prices began declining within a few quarters after May 2022, in conjunction with sharply rising mortgage rates, and that this downturn clearly showed up in the standard national housing statistics, matching Sacks’ prediction.

Evidence from mid‑ to late‑2022 shows that U.S. consumers experienced exactly the kinds of financial pressures Sacks described, within the “next several months” after May 2022:

• Higher borrowing costs. After the Fed began hiking in March 2022, consumer borrowing rates climbed quickly. By September–October 2022, average credit‑card APRs had risen from about 16.3% in March to over 18%, with analysts estimating cardholders would pay roughly $20–21 billion more in interest in 2022 than they otherwise would have. Mortgage and HELOC rates also jumped several percentage points over the year, materially increasing monthly payments for new or variable‑rate borrowers. (cnbc.com)

• Increased reliance on credit. The New York Fed’s Q2 2022 Household Debt and Credit report showed total household debt up to $16.15 trillion, with credit‑card balances alone rising by $46 billion in that quarter and up 13% year‑over‑year—the largest such increase in more than 20 years. (resources.newyorkfed.org) By Q4 2022, credit‑card balances had climbed to about $986 billion, surpassing the pre‑pandemic peak. (nasdaq.com) This indicates households were leaning more on revolving credit as prices and rates rose.

• Falling real wages. BLS data show that real (inflation‑adjusted) earnings were declining over exactly this period. Real average hourly earnings fell 3.0% over the 12 months ending May 2022, and even by late 2022 they were still down roughly 1.7–2.8% year‑over‑year, with real weekly earnings falling even more because average hours worked also slipped. (bls.gov) That’s a clear “reduced real wages” hit to consumers’ purchasing power.

• Drawdown of savings / thinner financial cushions. The U.S. personal saving rate dropped to about 2.7% in June 2022—its lowest level in roughly 15 years—and remained very low through late 2022 (around 2–4%), vs. a pre‑pandemic norm near 7–8%. (cnbc.com) This shows households largely stopped adding to savings and were instead using income and prior savings (and, as above, more credit) to maintain spending.

• Consumers as a key macro weak spot. The University of Michigan consumer‑sentiment index fell to 50.0 in June 2022, the lowest reading on record since the survey began in the late 1940s, reflecting sharply worse assessments of personal finances and buying conditions amid high inflation. (reddit.com) Subsequent commentary through late 2022 and into 2023 repeatedly highlighted strained household budgets and very weak sentiment as a central vulnerability, even as labor markets and corporate profits initially held up.

There is nuance: aggregate debt‑service ratios and pandemic‑era excess savings meant the consumer sector did not immediately collapse into a deep consumption‑driven recession, and overall spending remained surprisingly resilient for a time. (congress.gov) But Sacks’ specific mechanisms—higher borrowing costs, more use of credit, and declining real wages materially hurting consumers within months of May 2022, making households the next clear area of economic strain after the market sell‑off—did in fact materialize on schedule. On balance, the prediction is best judged as right.

What Chamath described as the Fed’s future QT path did not actually happen, even though the start date and initial plan he referenced were roughly right.

Key points:

1. What the Fed officially planned in May 2022

   • On May 4, 2022, the FOMC announced that balance‑sheet runoff (QT) would begin June 1, 2022, with monthly caps that ramp up to $60B in Treasuries + $35B in MBS = $95B/month, very close to Chamath’s “about $90 billion a month.” (forexlive.com)
   • So his description that QT would start around mid‑2022 at roughly $90B/month was aligned with the announced plan.

2. Actual pace was well below $90B/month on average

   • The Fed later slowed QT: on May 1, 2024 it reduced the monthly Treasury cap from $60B to $25B (while keeping MBS at $35B), cutting maximum monthly runoff from $95B → $60B starting June 1, 2024. (federalreserve.gov)
   • At the March 18–19, 2025 meeting it further cut the Treasury cap from $25B to $5B, leaving the MBS cap at $35B, so the max potential runoff dropped again to about $40B/month starting April 2025. (barrons.com)
   • These decisions mean the Fed clearly did not maintain QT at ~90B/month for three years; the average pace over the whole period is much lower.

3. Total runoff after three years was far short of $3T

   • According to the Fed’s own balance‑sheet report, as of March 26, 2025, securities holdings had declined about $2.1 trillion since QT began in June 2022. (federalreserve.gov)
   • A Congressional Research Service note (April 3, 2025) similarly states that the Fed had reduced its balance sheet by more than $2 trillion from its peak since starting QT in June 2022. (congress.gov)
   • Three years from the QT start (June 2022 → June 2025) therefore yields ~$2.1T–$2.2T of runoff, well short of Chamath’s “about $3 trillion.”

4. QT is ending with < $3T total reduction, even after more than three years

   • News reports on the Fed’s October 29, 2025 decision note that QT will end on December 1, 2025, with the balance sheet around $6.6T, down from a roughly $9T peak in 2022—i.e., about $2.4T of total reduction, not $3T. (reuters.com)
   • Even allowing for QT running longer than three years (June 2022 → December 2025 is ~3.5 years), the cumulative runoff still doesn’t reach $3T.

5. Comparison to Chamath’s explicit numbers

   • Chamath effectively assumed $90B/month × ~36 months ≈ $3.2T in runoff, summarized as “about $3 trillion” over “three years.”
   • In reality, policy changes slowed QT materially, delivering only ~$2.1T after three years and ~$2.4T by the time QT is scheduled to stop, so both the implied pace and the total were overstated.

Because the Fed did not maintain QT at roughly $90B/month, and total runoff over about three years (and even over the full QT program) falls well short of $3T, the prediction that it would take “about three years” at that pace to run off $3T of excess assets/liquidity is wrong as a forecast of what actually happened.

Evidence since May 2022 indicates that the U.S. did not enter an officially recognized recession, nor did it experience a sustained stagflationary episode.

• No official recession after 2020: The National Bureau of Economic Research’s business-cycle chronology still shows the last U.S. recession as the COVID-19 downturn, February–April 2020, with no additional peaks/troughs dated after that period. FRED and NBER’s own tables, which are updated as new recessions are declared, likewise list no post‑2020 recession dates. (nber.org)
• Growth remained positive overall: Real GDP grew about 2.5% in 2022, 2.9% in 2023, and 2.8% in 2024, indicating continued expansion rather than the broad, sustained contraction associated with a recession. (worldeconomics.com) While preliminary data once showed two negative quarters in early 2022, later revisions reduced this to a single clearly negative quarter (Q1 2022) followed by resumed growth, and mainstream analysis and NBER did not classify 2022 as a recession. (cnbc.com) Q1 2025 saw a small contraction (about –0.2/–0.3% annualized), but GDP then rebounded strongly in Q2 2025, again avoiding a sustained downturn. (investopedia.com)
• Inflation fell instead of persistent stagflation: Inflation was indeed high in 2022 (about 8% on an annual-average CPI basis), but then declined to 4.1% in 2023, 2.9% in 2024, and roughly 2.7% in 2025, returning close to the Federal Reserve’s target range. (usinflationcalculator.com) Over the same period, real GDP growth remained positive, contradicting the classic stagflation pattern of prolonged high inflation combined with flat or contracting output.
• Official and international assessments describe a soft landing, not stagflationary recession: IMF Article IV assessments and major economic commentary characterize the 2022–2024 period as one of resilient growth with inflation coming down, explicitly noting that the U.S. economy “dodged” or “skirted” the widely expected recession. (imf.org)

Friedberg’s prediction was that the U.S. was “inevitably going to run into a recession” in the near term due to stagflation. With the benefit of data through late 2025, the U.S. has not had a formally dated recession since 2020, growth has remained positive on an annual basis, and inflation has fallen substantially rather than staying elevated alongside weak growth. On standard economic definitions and official dating, the prediction is therefore wrong.

According to the Bureau of Economic Analysis (BEA), real U.S. GDP decreased at an annual rate of 1.6% in Q1 2022 (third estimate).(bea.gov) Subsequent BEA releases show that real GDP also decreased at an annual rate of 0.6% in Q2 2022 (third estimate).(bea.gov) The BEA’s own summary of the second estimate likewise states that real GDP decreased 0.6% in Q2 2022 following a 1.6% decrease in Q1 2022.(apps.bea.gov)

By the predictor’s explicitly stated technical definition—two consecutive quarters of negative real GDP growth—the U.S. was in a technical recession in Q2 2022. While the NBER did not later date an official recession for that period, the prediction was specifically about the technical criterion, which was met. Therefore, the prediction is right.

Evidence since mid‑2022 shows rising consumer credit and clear stress pockets, especially in credit cards and some auto and student loans, but not a full‑blown, systemic “ugly consumer credit bubble” that burst with a sharp correction, as described in the normalized prediction.

Key points:

• Rapid but not runaway growth in consumer borrowing. Federal Reserve G.19 data show total U.S. consumer credit outstanding rising from about $4.55T in 2021 to $4.89T in 2022 and $5.02T in 2023. Revolving credit (mainly credit cards) grew 15.1% in 2022 and 8.8% in 2023—fast, but not unprecedented.(federalreserve.gov)
• Record credit card balances but still a limited share of total debt. New York Fed reports show credit card balances hitting record levels (around $1.13T by Q4 2023, $1.14T by Q2 2024, and roughly $1.21T by Q4 2024 and into 2025).(newyorkfed.org) Even so, credit cards remain a relatively small slice of overall household debt compared to mortgages.
• Delinquencies did spike, but mainly in specific segments. By Q1–Q2 2024, about 8.5–9% of credit card balances and ~7.7–8% of auto loans were transitioning into delinquency annually, with serious credit card delinquencies (90+ days) reaching their highest level in over a decade (10.7% of card debt in Q1 2024).(newyorkfed.org) Lenders wrote off over $46B in delinquent card loans in the first nine months of 2024, the highest since 2010.(nypost.com) This is meaningfully stressful for lower‑income and younger borrowers, but it is not on the scale of a system‑wide credit collapse.
• Overall household delinquency rates remain moderate, and credit hasn’t crashed. Across all household debt, only about 3.2–3.6% was in some stage of delinquency in 2024, with the New York Fed repeatedly describing delinquencies as “elevated” but still near or below many pre‑pandemic norms, and noting that overall consumer financial health remains “solid.”(newyorkfed.org) There has been no sharp contraction in consumer credit outstanding; instead, total consumer credit has continued to edge higher through 2024 and into 2025, per G.19.(federalreserve.gov)
• No realized systemic crisis from consumer credit. Despite warnings from some commentators about a “credit bubble,” mainstream data and Fed communications do not indicate that consumer credit has created a systemic financial crisis akin to 2008. Household debt (even inflation‑adjusted) and overall delinquency rates remain below Great Recession extremes, and while subprime and younger borrowers are under strain, the banking system and broader economy have not undergone the kind of sharp, credit‑driven meltdown implied by a severe bubble thesis.(businessinsider.com)

Because the central features of the normalized prediction — a severe, systemic consumer credit bubble that then bursts in a sharp correction — have not materialized by late 2025, the prediction is best classified as wrong, even though it correctly anticipated rising balances and higher delinquencies in certain segments.

Available data from mid‑2022 through late‑2025 show no systemic or 2008‑style collapse in the overall U.S. real estate market, despite significant stress in some segments.

Key points:

• Home prices did not experience a 2008‑like crash. After the 2020–2022 boom, U.S. existing home prices largely plateaued or saw modest regional declines as mortgage rates spiked, but national price indexes (Case‑Shiller, FHFA) remained above pre‑pandemic levels and far above 2008–2012 troughs. Analysts repeatedly characterized the environment as a "housing slowdown" or "correction," not a systemic bust.
• Mortgage delinquency and foreclosure metrics stayed historically low. Even as rates rose and affordability deteriorated, serious delinquency and foreclosure rates remained well below Great Financial Crisis levels, largely because post‑2010 underwriting standards were tighter and homeowner equity was much higher.
• Banking system did not suffer a housing‑led cascade. The 2023 regional bank stresses (e.g., Silicon Valley Bank, Signature, First Republic) were linked primarily to interest‑rate risk and concentrated deposit bases, not to a wave of collapsing residential mortgages. Regulators and market commentary did not frame these as a repeat of the 2008 housing meltdown.
• Commercial real estate—especially offices—has been stressed but not systemically catastrophic. Rising rates and work‑from‑home trends produced sharp valuation declines and localized distress (notably in some downtown office markets). However, this has so far manifested as sector‑specific problems (refinancing stress, some bank/lender losses), not a broad‑based, economy‑threatening real estate crisis on the order of the 2008 nationwide housing collapse.

Given that by November 30, 2025 the U.S. has seen no major structural collapse of the overall real estate market comparable to the 2008 housing crisis, Chamath’s May 2022 prediction that "we don’t have an issue in real estate" in the sense of a 2008‑style systemic crisis is broadly borne out. The market cooled and certain segments (especially offices) have been under pressure, but the feared nationwide housing‑system meltdown did not materialize.

(Sources: aggregate reporting and data from Federal Reserve, FHFA, S&P CoreLogic Case‑Shiller, and major financial news outlets between 2022–2025 consistently describe a housing slowdown/correction and localized commercial‑real‑estate stress, not a systemic housing collapse like 2008.)

Key labor‑market data contradict both parts of Sacks’s prediction.

1. "Employment has peaked" by May 2022

• In May 2022 the labor market was still adding large numbers of jobs: nonfarm payrolls rose by about 241,000 that month, and then continued to increase every single month through the rest of 2022 and throughout 2023. Total nonfarm employment expanded further in 2024 (over 2 million additional jobs that year), showing that aggregate employment clearly peaked after 2022, not in May 2022. (bls.gov)
• BLS data for April 2023–April 2024 alone show payrolls rising from 155.5 million to 158.3 million, again demonstrating that employment levels were making new highs well past the prediction date. (bls.gov)

2. "Unemployment's going up" from the May 2022 lows in the ensuing months

• At the time of the episode, the U.S. unemployment rate was 3.6% in May 2022, the third consecutive month at that level. (virginiaworks.gov)
• In the following months, unemployment did not begin a sustained rise. It actually dipped to 3.5% in July 2022 (matching pre‑COVID lows) and was again 3.5% a year later in July 2023. (realestateguysradio.com)
• The jobless rate then fell further to 3.4% in 2023, a new roughly 50‑year low, meaning the post‑May‑2022 period included lower unemployment than at the time of his prediction, not an immediate uptrend. (apnews.com)
• A clear and lasting rise in unemployment only shows up later, with the national rate climbing into the low‑4% range by mid‑2024 and around 4.1% by early 2025—well after the “ensuing months” Sacks was talking about. (bls.gov)

Because (a) total employment continued to make new highs for years after May 2022, contradicting the claim that it had already peaked, and (b) unemployment first stayed flat/edges lower and even hit new lows before rising meaningfully much later, Sacks’s prediction is best classified as wrong.

The prediction was that, after roughly a 14% destruction of global wealth by May 2022, a large recession in the real economy would be inevitable and would materialize in the near term (1–2 years).

Key checkpoints between May 13, 2022 and May 13, 2024:

1. United States (most relevant real-economy benchmark)

   • The U.S. had two quarters of negative real GDP growth in early 2022, but that was before the podcast and was widely characterized as a “technical” or borderline case. The National Bureau of Economic Research (NBER), which is the standard arbiter for U.S. recessions, did not declare any U.S. recession for 2022–2024 up through mid‑2024.
   • U.S. real GDP continued to grow overall in 2022, 2023, and into early 2024, and the labor market remained historically tight with low unemployment and strong job creation. This is inconsistent with a “big recession” in the real economy.

2. Global picture

   • The IMF and World Bank documented a significant global slowdown, high inflation, and marked financial‑asset drawdowns post‑2021, but not a large, synchronized global real‑economy recession in 2022–2024. Instead, they described a period of “slow growth,” “stagflationary pressures,” and “fragmented” outcomes where some countries had mild or technical recessions, but the world economy as a whole continued to grow in real terms.
   • There were localized recessions (e.g., brief downturns in parts of Europe, the U.K., and specific emerging markets), yet these episodes were generally shallow or short‑lived rather than the sweeping “big recession” implied by the prediction.

3. Timing condition (1–2 years)

   • The prediction explicitly tied inevitability to the near term after May 13, 2022—i.e., by roughly May 13, 2024. By that point, despite substantial asset‑price declines and tighter monetary policy, the real‑economy outcomes were better described as a rough landing / soft-ish landing rather than a large, globally pervasive recession.

Because no major arbiter (e.g., NBER for the U.S., nor major international institutions for the world economy as a whole) characterizes 2022–2024 as a large real‑economy recession, and the dominant pattern was continued (if weak) growth plus localized, mostly mild recessions, the prediction that a big recession was “inevitable” in the subsequent 1–2 years did not come true.

Therefore, the appropriate classification is: wrong.

Chamath’s claim (May 2022) that the venture and growth equity ecosystem would go through “a few years of sorting”—with portfolio triage, layoffs, down rounds, and restructuring before reaching a new equilibrium—matches what actually happened from 2022 through at least 2024–2025.

• Multi‑year funding and valuation reset, not a quick snapback.

  • Global startup funding in 2023 fell sharply: Crunchbase data shows 2023 funding down 38% vs. 2022 and at the lowest level since 2018, with large pullbacks across all stages. (news.crunchbase.com)
  • Carta’s 2023 private‑markets data shows down rounds near 20% of all venture rounds in every quarter of 2023, the highest rates since at least 2018, signaling widespread valuation cuts and portfolio triage. (carta.com)
  • In 2024, PitchBook/Reuters report that global VC investment in Q1 2024 was at a near five‑year low, despite some big AI deals—evidence the broader market was still in a downturn nearly two years after his prediction. (investing.com)
  • Full‑year 2024 PitchBook–NVCA data shows global VC investment (~$368.5B) still down about 51% in value and 37% in deal count vs. the 2021 peak—a prolonged comedown rather than a 1‑year blip. (gayello.com)

• Portfolio triage, down rounds, and crossover/growth pullback.

  • A 2024 analysis of 2023 global venture reports notes that late‑stage valuations "are down massively from 2021" and that crossover funds like Tiger Global slashed activity (from 194 deals in 2021 to 20 in 2023) and tried to sell positions in secondaries at steep discounts, a clear sign of painful portfolio cleanup and growth‑equity retrenchment. (vccafe.com)
  • Carta and other data sources consistently show an elevated share of down rounds through 2023, and regional reports (e.g., Israel) indicate record levels of down rounds in 2024—26% of capital raises for mature companies—confirming that valuation resets and restructuring persisted into a third year. (carta.com)

• Layoffs and restructuring over multiple years.

  • The broader tech and startup ecosystem—which overlaps heavily with VC‑backed companies—has seen continuous large‑scale layoffs from 2022 through 2025. A 2025 report citing RationalFX data notes over 280,000 global tech layoffs in 2024, and another tally shows over 62,000 tech jobs cut in the first five months of 2025, with companies explicitly “right‑sizing” and restructuring after the pandemic‑era hiring boom. (axios.com)
  • Sectoral data (e.g., the video‑game industry) likewise documents significant layoffs every year from 2022–2025, illustrating that workforce reductions and cost‑cutting have been a sustained, multi‑year pattern rather than a brief shock. (en.wikipedia.org)

• Evidence of a new, more disciplined equilibrium emerging by 2024–2025.

  • A January 2025 growth‑equity update from Rothschild & Co notes that after three years of adaptation—shifting from “growth at all costs” to growth with profits/cash flow—valuation expectations are now tempered and company quality is better, making the universe of VC companies “more investable.” It also shows VC fundraising for funds in 2023–2024 far below 2021 and describes ongoing consolidation around larger managers, consistent with a reset to a new structure and equilibrium. (rothschildandco.com)
  • PitchBook–NVCA’s 2024/early‑2025 commentary similarly frames the environment as a subdued but stabilizing market, with dry powder, AI‑driven mega‑deals, and cautiously improving exits, but still far from 2021’s exuberance—exactly the kind of “sorted out” but lower‑temperature equilibrium he anticipated. (privatemarketsinsights.com)

Putting this together:

• The adjustment clearly lasted multiple years (2022–2024 and into 2025).
• It involved exactly the mechanisms he described: portfolio triage (e.g., secondary sales, markdowns), elevated down rounds, and waves of layoffs and restructurings.
• By 2024–2025, data and commentary describe a more disciplined, re‑priced, and consolidated market rather than a continuing free‑fall.

Given the timing and the qualitative shape of events, Chamath’s prediction that the venture and growth equity ecosystem would endure “a few years of sorting” before reaching a new equilibrium is well supported by the evidence, so the prediction is right.

Evidence since 2022 shows that venture funds, especially post‑boom vintages, have materially slowed deployment and shifted toward multi‑year investment periods, aligning with Sacks’ prediction.

1. Deployment pace clearly slowed after the 2020–2021 boom

   • Carta’s analysis of >2,000 VC funds shows that 2022‑vintage funds were only about 43% deployed after two years, versus ~50–60% for 2018–2020 vintages at the same age, and that the second year of deployment for 2022 funds was the slowest of any vintage since 2017. (carta.com)
   • Independent analyses (e.g., IDC Ventures’ Market Pulse) cite the same 43% figure and describe 2022 vintages as having “sluggish capital deployment,” confirming that capital is being put to work far more slowly than in the 2020–2021 period. (medium.com)

2. Industry commentary now explicitly points to 3‑plus‑year investment periods

   • A 2024 climate‑funding analysis, using Carta data, notes that 2020 funds deployed ~60% of capital within 24 months while 2022 funds deployed only ~43%, and concludes that funds announced in 2023–24 are trending toward 3–4‑year investment periods—a direct confirmation that the “one‑year pace” of the boom has given way to a multi‑year cycle. (slideshare.net)
   • A 2025 overview of U.S. VC notes that the 2020–2022 fundraising surge left managers with substantial dry powder, and that weaker exits and LP pullbacks in 2023–2024 forced funds—especially large, established ones that now receive ~80% of capital—to invest more cautiously over longer periods instead of rushing into deals. (afurrier.com)

3. Fundraising cadence for big, traditional firms also lengthened, consistent with slower deployment

   • An analysis of U.S. venture fund cycles finds that during the 2020–2021 boom many firms raised new funds every 10–15 months, but by 2024 the median gap between successive funds had stretched to ~24 months (with most between 20 and 34 months). This doubling of time between funds is consistent with managers no longer deploying entire vehicles in roughly a year. (unlistedintel.com)
   • Carta’s 2025 fund‑performance data show that even rebounding 2023 vintages are only about 58% deployed after six quarters, still implying a multi‑year path to full deployment, not the near‑one‑year cycles seen at the peak. (linkedin.com)

Taken together, these data and commentaries show that the rapid one‑year deployment behavior of 2020–2021 has largely ended and that large, traditional venture funds are again operating on roughly three‑year (or longer) deployment cycles. That matches the substance and direction of Sacks’ prediction, so it is best judged as right.

Evidence from 2022–2024 shows that the VC downturn Sacks anticipated did occur, and that investors systematically favored high‑growth, capital‑efficient startups while pulling back hardest from high‑burn, mediocre‑growth companies.

• Clear downturn in venture funding starting in 2022. Global VC funding fell ~35% in 2022 vs. 2021 and then another ~42% in 2023, reaching the lowest level in years, with deal counts also dropping sharply.(livemint.com) This created exactly the “downturn” environment the prediction was about.
• Investor criteria shifted to “efficient growth” (high growth, moderate burn). VC commentary and guidance from 2022 onward consistently emphasize capital efficiency, burn multiple, and Rule of 40 over “growth at all costs.” TechCrunch in April 2023 described a “tectonic shift,” saying capital efficiency had become the new filter and that companies bleeding cash with weak efficiency metrics would be deemed “default dead” or face sharply reduced valuations.(techcrunch.com) Other analyses and VC guides explicitly note a post‑2022 move away from rewarding pure top‑line growth toward sustainable, efficient growth and strong unit economics.(raisebetter.capital) A 2025 overview even cites David Sacks’s burn multiple as a now‑standard metric and uses Notion’s low burn multiple as a reason it could attract “massive funding rounds even in tight markets” – a direct example of high growth with moderate/efficient burn continuing to get funded.(medium.com)
• High‑burn, modest‑growth startups repeatedly failed to raise and were forced to shut down or restructure. CB Insights’ large compilation of startup failure post‑mortems documents many 2022–2023 closures where unsustainable burn plus insufficient growth meant new funding could not be raised in the tighter VC market. For example, Fast (one‑click checkout) had burn “far exceed” revenue growth; when the market turned, it couldn’t raise more capital and shut down. Rapid‑delivery and other high‑burn models similarly folded or sold in 2022 as funding tightened.(scribd.com) Broader reports on the “funding winter” note tens of thousands of startup closures in 2023, with many citing high burn, lack of clear profitability, and inability to secure follow‑on capital as key causes.(livemint.com)
• Meanwhile, selective exceptions with extreme growth still got funded, not contradicting the rule. Even within the funding winter, sectors like generative AI raised record sums while overall startup funding was down ~65% from 2021, illustrating that investors still backed very high‑growth stories, but in a far more selective way.(globaldata.com) Those winners generally also stressed efficiency and strong business metrics, consistent with the idea that “high growth + reasonable burn” remained fundable while weak or moderate growth + heavy burn became untenable.

Taken together, the data and post‑mortems show that after 2022, venture capital markets did exactly what Sacks predicted at a stylized level: during the downturn, funding concentrated in companies showing strong growth with controlled burn and solid unit economics, while many startups with only moderate growth and high burn were unable to raise, and were forced into shutdowns, fire‑sales, or drastic cost‑cutting. The prediction is therefore best characterized as right in its core claim about how funding would bifurcate during the 2022+ downturn.

Evidence since 2022 shows that the big crossover hedge funds did sharply pull back from new private tech investing for several years, and they stopped being the dominant marginal capital in late‑stage rounds, which is the essence of Sacks’s prediction.

Key points:

1. Crossover activity collapsed post‑2021
   A 2023 VC review notes that major crossover funds such as Tiger Global, Temasek, Coatue, and SoftBank went from participating in $233 billion of VC deals in 2021 to just $34 billion in 2023. It characterizes this as a “retrenchment” that had a notable impact on late‑stage funding, with corporates and sovereign wealth funds “picking up the slack” as crossover funds scaled back. (allianceequityresearch.com)
   An India‑focused funding analysis similarly describes ~90% compression in crossover‑fund activity, explicitly citing Tiger Global and SoftBank as having “significantly reduced deal‑making.” (tice.news)
   This is consistent with “effectively withdrawing” as dominant late‑stage capital providers.

2. Tiger Global specifically pulled back and shrank its venture footprint
   Coverage of Tiger’s strategy after its 2022 losses says it “actively slowed its venture investments in 2022, curtailing their size and number,” and shifted away from later‑stage growth equity toward earlier‑stage VC bets. (thewealthadvisor.com)
   When Tiger finally closed its next private fund (PIP XVI) in 2024, it was only $2.2 billion, far below the $6 billion target and a fraction of its $12.7 billion 2021 fund, underscoring how much its late‑stage firepower had diminished and how cautious LPs had become. (ft.com)
   This supports the idea that Tiger was not rapidly redeploying massive late‑stage capital after 2022.

3. Coatue dramatically reduced deal volume and shifted strategy
   Crunchbase data show that Coatue’s venture deal count fell from 168 deals in 2021 to just 29 in 2023 (an 82% decline), and the total dollar amount of those deals dropped from about $43 billion to $4.1 billion. (news.crunchbase.com)
   While Coatue remained active in some large structured or AI‑related financings (e.g., leading a $1.1 billion round for AI cloud startup CoreWeave in 2024), these are exceptions within a much smaller, more selective portfolio—consistent with a retreat from the broad, aggressive late‑stage strategy that defined the 2020–21 boom. (marketwatch.com)

4. D1 and other crossovers also refocused away from new late‑stage privates
   Reporting on D1 Capital and similar crossover funds notes that after the tech sell‑off, D1 explicitly told startups it was slowing new private investments and instead was reallocating toward beaten‑down public names. (theinformation.com)
   A broader 2025 industry overview summarizes that when markets turned in 2022, many crossover investors pulled back dramatically from private tech investing, particularly at late stage; crossover capital remained in play only in a more limited, selective way. (afurrier.com)

5. Late‑stage capital shifted back to traditional VC/PE and corporates
   With crossover funds scaling back, late‑stage companies increasingly relied on large, traditional VC and growth‑equity firms and on corporates/sovereign wealth funds. The same 2023 review explicitly says corporates and SWFs “picked up the slack” as crossover funds retreated. (allianceequityresearch.com)
   Separately, PitchBook data cited by the Financial Times show 2024 U.S. VC fundraising heavily concentrated in a handful of big Silicon Valley firms (e.g., Andreessen Horowitz, General Catalyst, Thrive, Iconiq), which jointly raised the majority of VC dollars, reinforcing that traditional VCs became the main organized capital pool while crossover funds were nursing losses and retrenching. (ft.com)

6. They did eventually come back—especially in AI—but after a multi‑year lull
   By late 2024–2025, crossover names like Tiger Global and Coatue reappear in marquee AI rounds (e.g., Tiger planning to join a multibillion‑dollar OpenAI round; Coatue leading or joining billion‑dollar financings for CoreWeave), and Coatue is raising new AI‑focused capital. (reuters.com)
   However, that resurgence followed roughly two to three years (mid‑2022 through much of 2024) of sharply reduced, selective activity compared to their 2020–21 role. During that period, they were no longer the primary marginal capital in late‑stage private tech.

Putting this together: crossover hedge funds didn’t disappear entirely, but they did substantially withdraw from aggressive late‑stage private tech investing for several years after the 2022 correction, and other investors—especially large traditional VC/growth funds plus corporates/SWFs—became the main sources of capital in that segment. That outcome aligns closely with the normalized version of Sacks’s prediction, even if his on‑air phrasing (“they’re gone”) was somewhat hyperbolic and they eventually returned in a narrower, AI‑focused way.

As of November 30, 2025, no new technology company that emerged from the roughly $250 billion in VC dry powder available around mid‑2022 has grown to a $1 trillion+ public market capitalization.

Public members of the trillion‑dollar “mega‑cap” club in 2025 include Apple, Microsoft, Nvidia, Alphabet, Amazon, Meta, Tesla, Broadcom, TSMC, Berkshire Hathaway, Saudi Aramco, and a few others, all of which are long‑established firms founded well before the 2010s and certainly before the 2020–2022 VC boom. (bankrate.com) A recent example outside core tech, Eli Lilly, also hit $1 trillion in 2025 but is a 19th‑century company, not a new VC‑backed startup. (barrons.com) There is no evidence of any startup founded or primarily funded in the 2020–2022 period that has reached a $1 trillion valuation by late 2025.

However, Friedberg’s statement doesn’t specify a clear time horizon; he says only that "one of the next trillion dollar mega‑caps" could emerge from that roughly quarter‑trillion of VC capital and, if it did, the industry "would look fantastic" and that company would "transform the landscape of some part of the economy." VC fund cycles and company‑building timelines are typically on the order of a decade or more, not three years, so by late 2025 it is still too early to say definitively that no such trillion‑dollar company will emerge from that capital.

Because (1) the necessary condition (a new VC‑backed tech company from that capital reaching a $1T+ market cap) has not yet occurred, but (2) the prediction has no explicit deadline and venture timeframes plausibly extend well beyond 2025, the status of the prediction is best classified as inconclusive (too early to tell) rather than clearly right or wrong.

The prediction specified that within ~9–12 months of mid‑May 2022 (i.e., by around May 2023) the U.S. would hit a consumer credit bubble that turned into a visible credit crisis, where people would broadly question whether consumers could afford their accumulated credit card and loan debt.

What actually happened by that time:

• The New York Fed’s Q1 2023 Household Debt and Credit report (covering data through March 31, 2023) shows total household debt at $17.05T, with credit card balances essentially flat that quarter at about $986B. Transitions of credit card balances into serious delinquency (90+ days late) did rise year‑over‑year (from 3.04% in Q1 2022 to 4.57% in Q1 2023), but overall serious delinquency rates remained low, and new foreclosures stayed very subdued. The Fed did not characterize conditions as a crisis. (newyorkfed.org)
• A February 16, 2023 CNBC summary of the New York Fed’s Q4 2022 data reported record consumer debt of $16.9T and noted that mortgage, auto, and credit‑card delinquencies had increased “though to still‑low levels,” again indicating rising stress but not a credit crisis. (cnbc.com)
• The Q2 2023 New York Fed report (released August 8, 2023, covering data through June 30) highlighted that credit‑card balances crossed $1T and delinquency transition rates were rising, but it framed this as gradual normalization from unusually low pandemic levels, not a full‑blown crisis in which consumer credit quality was collapsing. (newyorkfed.org)
• Commentary explicitly using “consumer credit bubble” language and warning that higher rates were threatening such a bubble appears later in 2023; even then, analysts describe elevated risks and rising defaults rather than a discrete, already‑materialized crisis by mid‑2023. (hva.group)
• The large, widely reported jump in serious credit‑card delinquencies—more than 50% year‑over‑year—shows up in full‑year 2023 data that the New York Fed and outlets like CNBC discussed in early 2024, again framed as increased financial stress, especially for younger and lower‑income borrowers, not as a systemic credit seizure akin to 2008. (cnbc.com)

By the end of the stated window (around May 2023), U.S. consumer credit conditions were tightening and deteriorating from unusually strong post‑pandemic levels, but there was no widely recognized consumer credit crisis centered on household inability to service credit card and loan debt. The more severe deterioration and “crisis”‑type language emerge later and are still typically described as elevated stress, not a discrete crash event. Therefore, judged against both the timing and the severity/"credit crisis" framing in the prediction, it did not come true.

On the episode dated 13 May 2022, the S&P 500 closed at 4,023.89, with recent lows in the 3,800s that week, so Chamath’s “3800‑ish” reference matches the then-current trading range.

How far did the S&P 500 ultimately fall?
• The 2022 bear‑market closing low was 3,577.03 on October 12, 2022, and the intraday low was 3,491.58 on October 13, 2022. (statmuse.com)
• That puts the trough roughly 6% below 3,800 on a closing basis and ~8% below on an intraday basis—well above the ~3,000 level he discussed as a theoretical valuation floor.

Time spent “significantly below” 3,800
The index did trade below 3,800 for stretches (notably mid‑June and late September–mid‑October 2022, with closes in the 3,600s and high‑3,500s), but by December 30, 2022 it had recovered to 3,839.50, back above 3,800. (indices.earningsahead.com) Those sub‑3,800 periods totaled a few weeks at a time, not a deep, multi‑quarter stay far below that level.

Did the market ever approach 3,000?
No. Across 2022 the S&P 500’s year low was about 3,577 on a closing basis, roughly 25% below its January 3, 2022 high of 4,796, never coming close to 3,000. (en.wikipedia.org) After that, the index entered a new bull phase, making new highs in 2023–2025 above 4,500, 5,000, and then 6,000+. (sec.gov)

Assessment vs. the prediction

• Core claim: "we're probably close to a near bottom‑ish here, 3800‑ish" – The actual bottom in the mid‑3,500s was only single‑digit percent below that level, which is reasonably "near" in macro terms.
• Scenario analysis: He said valuations could justify ~3,000 but expected a policy/market response (“Fed put somewhere in between here”) to prevent that. The realized low around 3,500 is indeed between 3,800 and 3,000 and never approached 3,000.

While the index did go modestly and temporarily below 3,800, the eventual trough and subsequent path line up with his main directional call: 3,800 was close to the cyclical bottom, and the feared move toward 3,000 never happened. On balance, that makes the prediction substantially correct.

Available data are mixed and the time horizon is probably too short to judge this strong, long‑run macro claim.

On the deployment of dry powder:

• U.S. VC firms had about $311.6B in dry powder at the end of 2023, more than the “~$250B” cited in mid‑2022, indicating that uncalled capital has grown, not been substantially run down.(nvca.org)
• Globally, PE+VC funds held a record $2.62T of uncommitted capital as of July 2024, with a rising share of “aging” (4+ year‑old) capital, suggesting significant difficulty deploying existing dry powder.(forbes.com)
  This contradicts the idea that most of the mid‑2022 dry powder has already been actively deployed in the subsequent few years.

On actual VC investment levels:

• U.S. VC investment fell from $242.2B in 2022 to $170.6B in 2023, and global VC fell from $435.7B in 2022 to $260B in 2023, a ~40% drop, consistent with a broad pullback.(spglobal.com)
• In 2024, there was some rebound: U.S. VC funding rose to about $209B, with AI startups taking a record ~46% share, and global VC rose modestly to $314B, still ~55% below the 2021 peak.(reuters.com)
  So capital is being deployed, but not at clearly extraordinary levels relative to the pre‑bubble trend, and not in a way that visibly exhausts the 2022 dry‑powder overhang.

On real‑economy impact (jobs and new growth areas):

• Historical work shows venture‑backed firms are important for jobs and GDP, but those studies mostly cover periods up to ~2020 and don’t isolate the post‑2022 dry‑powder deployment.(ventureforward.org)
• There are localized signs of VC‑linked job creation (e.g., Virginia’s 10,000+ new high‑growth, high‑wage startups in 2022–2023, supported in part by rising VC investment in the state), but these are not enough to conclude that, in aggregate, new VC‑funded firms have already “offset much of” the speculative excess of the 2020–2021 bubble.(axios.com)
• The post‑2022 period also saw major tech layoffs and a sharp slowdown in startup funding outside of hot areas like AI, which cuts against a clear narrative of broad, net‑new employment gains driven by that specific pool of dry powder.(spglobal.com)

Because:

1. a large share of the dry powder present in mid‑2022 is still uninvested by 2024–2025;
2. the deployment that has occurred is heavily concentrated (e.g., in a few large AI rounds) rather than clearly broad‑based; and
3. it is inherently hard, after only ~3 years, to measure whether this capital has “offset much” of prior speculative excess via net job creation and new growth sectors,

there is not enough evidence yet to decisively label the prediction as either right or wrong. The fairest assessment as of late 2025 is that it is too early to tell, i.e., inconclusive.