Coatue's 'Hard Truth Index': Is the 2025 market a correction or a crisis?
Jun 3, 2025
Key Points
- Coatue's Hard Truth Index shows market sentiment at negative 1.4 while economic data sits neutral at negative 0.3, suggesting 2025 is a sharp correction rather than crisis based on year-to-date earnings down only 5% versus 20%+ in past crises.
- OpenAI reached 800 million users within weeks of launching Deep Research and Operator, doubling from 400 million and outpacing Twitter's 96-month climb to 288 million, while Microsoft processed 100 trillion tokens last quarter up 5x year-over-year.
- Tech valuations compressed to 24x forward earnings versus 18x for non-tech, down from a 9x spread at peak AI boom, yet founder-led companies like Tesla and Nvidia that survived 2022 rate-shock corrections historically emerge stronger through faster strategy reorientation.
Summary
Coatue's public market update frames 2025 as a critical inflection point between correction and crisis, a distinction with material implications for venture capital and tech valuations.
The Hard Truth Index
Coatue compares economic hard data to market sentiment using a quantitative framework. Sentiment has spiked to negative 1.4 while hard economic data sits at neutral (negative 0.3). This mirrors what some call the "vibe session"—a period where market psychology and actual business conditions severely misalign.
Past crises leave unmistakable marks. The dotcom crash saw a 133-week drawdown and 49-50% peak-to-trough decline in the S&P 500, accompanied by 42% earnings cuts. The 2008 financial crisis matched that severity with a 74-week drawdown, 57% peak-to-trough decline, and 42% earnings cuts. COVID was sharper but shallower—five weeks of drawdown, 34% decline, and a 21-week recovery.
Corrections are shorter and shallower. The 2022 inflation correction saw a 40-week drawdown and 25% peak-to-trough decline, but earnings fell only 1%. Duration and earnings impact distinguish the two.
Where 2025 Sits
The year's first four months delivered an 18% drawdown, the worst start since at least 2001. Yet current signals point toward correction rather than crisis. Year-to-date earnings are down just 5%, well below the 20% plus threshold for a true crisis. The VIX spike was brief, peaking at 52 before falling to 23. Two key circuit-breakers arrested the decline. A 90-day tariff pause announced April 8 marked the market bottom. A regional Fed president signaled willingness to add temporary liquidity if tariffs continued to worsen.
Consumer data strengthens the case. Visa, Capital One, Bank of America, and American Express all reported in late April that discretionary and non-discretionary spending remain resilient. Stock market moves have not deterred spending. Restaurants in wealthy coastal areas report packed lunch crowds despite doom-heavy headlines.
The AI Overhang
AI adoption is accelerating while macro uncertainty swirls. OpenAI reached 400 million users roughly 24 months post-launch, then jumped to 800 million after releasing Deep Research and Operator—a doubling in weeks that Sam Altman confirmed. Twitter reached 288 million over 96 months. Instagram and Facebook each took years to hit a billion.
Microsoft reported processing 100 trillion tokens in the quarter, up 5x year-over-year, with a record 50 trillion tokens in a single month. This speaks to explosive inference demand, a shift from training-heavy to inference-heavy workloads that will shape GPU demand going forward. Nvidia has become deliberately less transparent about the training-to-inference ratio, making that calculation a key unknown for the coming quarters.
Valuation Reset
Tech multiples have compressed but remain elevated relative to non-tech peers. The Magnificent 7 grew revenue at 30% annualized in Q1 while the remaining 493 Fortune 500 companies grew at 8%. Tech trades at 24x forward earnings versus 18x for non-tech, a 6x spread down from the 9x spread during the peak AI boom.
Some franchise names have been hit hard. Reddit down 49%, Tesla down 40%, ARM down 34%, Shopify down 23%, Nvidia down 23%. These are founder-led or founder-aligned businesses that have weathered prior corrections by moving faster than incumbents. The pattern suggests that during rate-hike shocks and macro turbulence, founder-led tech companies that can reorient strategy quickly emerge stronger.
The Core Question
Coatue concludes it does not seem like a crisis yet, more like a sharp correction. The implication is clear—someone must be wrong, and sentiment appears disconnected from reality.
Historically, extreme bearishness alone is not a reliable buy signal. When business sentiment hits the 10th percentile, the S&P 500 delivers top 20% one-year returns only 35% of the time. The bottom 20% of returns occur 8% of the time. Nearly 57% of the time, the outcome is unknowable.
Coatue's stated conviction: tokens matter more than tariffs. If true, the macro noise around trade policy becomes a sideshow to the underlying AI infrastructure buildout, one that is visibly accelerating despite headline turbulence.