The Numb Market

On February 20, 2026, the Supreme Court ruled 6-3 that the International Emergency Economic Powers Act does not authorize the reciprocal tariff regime that produced the entire Liberation Day shock, voiding the statutory foundation under most of the 2025 trade architecture.¹ The S&P 500 closed up 0.69% on the day. Two weeks later, on March 4, 2026, Iran closed the Strait of Hormuz, and the IEA classified the resulting disruption as the largest in oil market history, with global supply falling 10.1 million barrels per day.² The S&P 500 closed April at an all-time high of 7,230.⁸ In the same window, the University of Michigan's Current Economic Conditions sub-index posted the lowest reading on record going back to 1952.³ These four data points belong to different macroeconomic regimes, and all four are current as of this writing.

The bull case has real earnings behind it. Q1 2026 blended EPS growth is tracking 27.7%, the highest since Q4 2021. Net profit margins are at a record 13.4%, and 84% of S&P 500 companies beat earnings estimates, the strongest beat rate since Q2 2021.⁴ None of that is in dispute. What is also true, and harder to fit into the same frame, is that the market has built a tolerance to Trump. Every shock between Liberation Day and February 2026 was followed by a reversal, a court intervention, or a data print that rewarded buyers, and after sixteen months of that pattern the market stopped selling on news.

This report makes two arguments. The first is that equity markets have undergone a measurable habituation process in response to erratic White House policy signaling, with the informational content of each successive shock approaching zero. The second is that the causal arrow often runs in reverse: Trump treats markets as a behavioral feedback signal, escalating when equities are calm and retreating when bonds move. The combination is what makes the current regime fragile. A market that no longer prices Trump has lost the ability to transmit the corrective signal he reads in deciding when to back off, which means risk accumulates while the dashboard he watches stays green.

There is a paradox at the center of the Trump economy. No president in modern history has generated more policy shock per day: tariffs announced, reversed, expanded, paused, re-announced; trade wars declared and walked back within weeks; executive orders signed and then voided. In February 2026 the Supreme Court struck down the statutory basis of the reciprocal tariff regime itself.¹ And yet, as of May 2026, the S&P 500 sits at an all-time high.

In Afghan folklore there is a story called Choopan-e Dorough-go (چوپان دروغ‌گو), the lying shepherd. A boy on a hillside outside the village cries "wolf" for sport. The villagers come running and find nothing. When the wolf finally arrives, he is met with silence. Economist Előd Takáts named this dynamic in a 2007 IMF paper: crying wolf economics, the case that repeated false alarms eventually degrade a warning system until the real signal cannot be heard.

Takáts was writing about money laundering reporting, but the mechanism generalizes, and it has a clinical name in psychology. What behavioral economists call habituation, a term Richard Thompson and Alden Spencer formalized in 1966 as the process by which a repeated stimulus produces a diminishing response, is the literal mechanism at work in the market. In market terms the village is still standing. The structural risks (fiscal imbalances, consumer bifurcation, geopolitical exposure in the Persian Gulf, a depreciating dollar) have not been resolved. They have been priced out of the forward curve because the market has stopped listening, and a market that has stopped listening has stopped functioning as the warning system it was supposed to be.

The trajectory of market response to Trump policy signals over the past sixteen months traces a habituation arc, with each successive shock producing a smaller market reaction than the one before it. The curve flattens phase by phase, and by the fourth phase it has flattened to roughly zero.

Phase 1, the optimism rally (January 2025). Markets rallied into inauguration day as investors priced in a pro-growth administration: tax cuts, deregulation, AI investment tailwinds. Early tariff threats on Colombia, then Mexico and Canada, were dismissed after quick reversals. The interpretive pattern was set early, with threats read as negotiating theater rather than policy, and the S&P 500 hit a new all-time high above 6,144 in February 2025.

Phase 2, the first habituation (February to March 2025). As tariff announcements multiplied, markets initially repriced Trump's protectionist rhetoric as negotiating strategy. Repeated reversals eroded trust on both sides. Volatility rose, and market participants learned the rhythm: buy the dip, wait for the reversal.

Phase 3, Liberation Day and the last real panic (April 2 to 9, 2025). On April 2, 2025, Trump announced sweeping reciprocal tariffs from the Rose Garden, including 34% on China, 20% on the EU, and 46% on Vietnam. Over April 3 and 4, the S&P 500 fell 10.5%, the worst two-day decline since the pandemic crash of March 2020.⁶ The VIX hit 60.13 intraday on April 7, its highest level since the COVID shock,⁵ and the index touched 4,983 on April 8, briefly entering bear-market territory. Roughly $10 trillion in global equity value was erased in a week.

What ended the panic was not equity losses, which Trump publicly dismissed. The 10-year Treasury yield moved from a 3.87% low on Monday April 7 to 4.47% on Wednesday April 9, a 60 basis point three-day swing that was the largest since December 2001.¹⁴ The 30-year posted its sharpest move since the 2020 pandemic shock. On April 9, Trump paused tariffs on most nations, citing the bond market directly:

Phase 4, the new regime (May 2025 to present). From May 2025 onward, markets entered a regime of structural indifference to headline policy shock. The S&P 500 returned 17.9% total return in 2025, surged nearly 39% from the April low through year end, and closed April 2026 at an all-time high of 7,230.⁸ The Nasdaq Composite closed April 2026 at approximately 24,892. Investors rotated their analytical framework, moving away from "will this announcement cause volatility?" toward "will this permanently impair earnings and growth?" With Q1 2026 blended earnings growth tracking 27.7%, the answer was repeatedly no.⁴

What was a 10% drop in April 2025 became a 1% drift in February 2026 and roughly zero by March, even as the underlying policy shocks continued to roll in at roughly the same scale. The chart is tracking that change in the market, not a change in the policy environment.

Three reinforcing mechanisms drive market habituation, making each successive shock land smaller than the last:

1. Buy-the-dip conditioning. Every major shock between January 2025 and February 2026 was followed by a policy reversal or a fundamental data print that rewarded buyers. The brain of the market, meaning the aggregate behavior of fund managers, quant models, and retail traders, learns through experience. When selling is punished repeatedly and buying is rewarded, the market stops selling. The April 9 reversal was the formative event: a 9.5% one-day rally on the S&P 500, the largest single-day gain since 2008 and the third largest of the postwar era,⁹ which trained every fund manager who had sold on Liberation Day to wait for the off-ramp instead. The lesson, internalized, was that panic is expensive. Conditioning built on reversals assumes reversals will continue, which becomes dangerous the first time one does not come.

2. Fundamental override. Market desensitization was reinforced by truly strong fundamental data. S&P 500 Q4 2025 EPS growth landed at roughly 14% year over year, and Q1 2026 came in at 27.7% blended growth, the highest since Q4 2021. The 84% beat rate of Q1 2026 was the strongest since Q2 2021, and net profit margins of 13.4% are the highest FactSet has tracked since the series began.⁴ AI capex was the primary engine: the five largest hyperscalers spent roughly $437 billion in capital in 2025, up about 62% year over year and nearly 2.5 times 2023 levels, and 2026 hyperscaler capex has been revised up 12% to roughly $720 billion.¹⁰ Total S&P 500 capex grew 39% year over year in Q1 2026.¹¹ In a world of strong earnings, political noise is discountable. The rational investor asks whether a given tariff or executive order will materially change the earnings outlook of the companies they own, and for most of 2025 and into 2026 the answer was no. Fundamental override is a legitimate analytical posture, but it creates a dangerous asymmetry the moment fundamentals begin to soften.

3. Information decay. Each successive Trump event generates less new information for markets to process. The market's prior distribution already contains the possibility of erratic announcements, so when an event occurs it confirms an expectation rather than creating a new one, and the informational content of the shock approaches zero. The cleanest empirical demonstration is the SCOTUS ruling. On February 20, 2026, the Supreme Court ruled 6-3 that the IEEPA did not authorize the tariff regime that delivered the entire Liberation Day shock. Hours later, Trump announced a 10% global tariff under Section 122 to replace it. JPMorgan's trading desk had assigned a 64% probability to exactly this outcome (struck down and replaced), and the S&P 500 closed up 0.69%.¹ The market knew before the gavel. If every announcement is reversible, no single announcement is news.

The conventional narrative of the Trump-market relationship is one-directional: Trump announces policy and markets react. The more accurate picture is bidirectional, and the feedback running from markets back to Trump is the more consequential, and less covered, half of the dynamic. Habituation breaks this loop on the receiver's side. A market that has stopped reacting has also stopped sending the only signal that has historically corrected the policymaker.

The April 9, 2025 tariff pause is the cleanest natural experiment available. Trump had publicly dismissed equity market losses for days, calling them temporary and irrelevant. What he could not dismiss showed up in the bond market.

As Treasury yields surged, with the 10-year posting its largest three-day jump since 2001 and bond auctions deteriorating, Treasury Secretary Scott Bessent and chief of staff Susie Wiles began fielding calls from business executives. JPMorgan CEO Jamie Dimon told Maria Bartiromo on Fox Business that recession was "a likely outcome." Within hours, the pause was announced via Truth Social.¹⁵ National Economic Council director Kevin Hassett later told CNBC that "the fact that the bond market was telling us, hey, it's probably time to move, certainly would have contributed at least a little bit to that thinking."¹⁶

Trump later told Time magazine, "I wasn't worried. The bond market was getting the yips, but I wasn't."¹⁵ The denial is itself the tell, because no single number forced his hand. Equity losses were real and politically expensive, yet he could stand in front of cameras and call them temporary; what he could not wave off was everything arriving at once. The bond selloff and the weak auctions were landing on a national debt measured in the tens of trillions, where every additional basis point of yield is real money in added interest cost, and they came alongside executive phone calls and open recession warnings from figures the market actually listens to. He moderates when several pressures point the same way at the same time, not when any one of them does on its own.

Which makes the present moment the strange one: every signal in that cluster is now sitting where it would once have forced a response, and none of it is landing. The dollar closed 2025 down 9.6%, its worst year since 2017 and its worst first half in more than fifty years;¹⁸ recession indicators are flashing, a shooting conflict in the Persian Gulf is squeezing the global energy supply, and equities are still at record highs, yet a combination that forced a policy reversal within days a year ago now produces almost nothing. The mechanism that used to correct him has not disappeared; it has gone quiet.

The danger in the current environment is not a repeat of Liberation Day volatility but the opposite of it. Volatility, paradoxically, is a feature of a functioning market, since it signals that risk is being priced and communicated; the regime to worry about is a calm market sitting atop a deteriorating fundamental picture, where the absence of alarm is itself the alarm. Sixteen months of habituation have produced exactly this regime, and the list of items the market is currently not pricing is long:

A. Consumer bifurcation. Headline consumer spending figures mask a severe bifurcation. The top 10% of earners account for nearly half of all U.S. consumer spending,²¹ which means aggregate data is a wealth-weighted average that systematically overstates the health of the broad economy. Meanwhile, consumer sentiment has collapsed. The University of Michigan Consumer Sentiment Index fell to 56.4 in January 2026 and continued to slide, hitting a preliminary 48.2 by May, below the level recorded at the start of every U.S. recession since the series went monthly in 1978. The Current Economic Conditions sub-index hit 47.8, the lowest reading on record in the survey's history dating to 1952.³ The Conference Board Consumer Confidence Index dropped to 84.5 in January 2026, its lowest since 2014.¹³ About a third of UMich respondents now spontaneously mention gasoline prices and roughly 30% mention tariffs, and the declines are bipartisan across age, income, and political affiliation.³

The bifurcation between equity markets near all-time highs and consumer confidence at record lows is the central anomaly of the current economic moment, and it resolves in one of two directions. Either sentiment recovers as inflation cools and labor markets hold, or spending eventually contracts as the psychology of the median household catches up with behavior. Strong corporate earnings and record-low consumer sentiment are not stable companions for long, and whichever side of the bifurcation resolves first will determine the trajectory of the rest.

B. The structural fiscal picture. The federal budget deficit was 5.9% of GDP in FY2025 (CBO), totaling $1.8 trillion. Gross interest costs surpassed $1 trillion for the first time, and net interest of roughly $970 billion became the second-largest federal expenditure behind Social Security. The CBO projects deficits in the 5.6% to 6.5% of GDP range through 2035, with debt held by the public crossing 100% of GDP in 2026 and reaching 118% to 120% by 2035.¹⁹ The Yale Budget Lab puts the average effective tariff rate at 11.0% in April 2026, the highest since 1943; Penn Wharton has the figure at 8.9% through February 2026.¹⁷, ²⁰ Even on the lower estimate, this is the most tariff-heavy U.S. trade regime since 1947. These are structural pressures rather than cyclical headwinds, and a market habituated to ignoring Trump's policy signals is also a market at risk of underpricing the cumulative effect of those policies on the fiscal trajectory.

C. The Iran war and the Hormuz shock. U.S. and Israeli airstrikes against Iran resumed on February 28, 2026, escalating a confrontation that began with the June 2025 U.S. strikes on Iranian nuclear facilities at Fordow, Natanz, and Isfahan.⁷ Beginning March 4, 2026, Iranian forces declared the Strait of Hormuz closed, with at least 10 confirmed attacks on transit vessels in the first four days. Roughly 27% of global maritime crude and petroleum products trade transits the Strait, and the IEA classified the resulting disruption as the largest disruption in history. Global oil supply fell 10.1 million barrels per day in March to 97 mb/d, with OPEC+ output falling 9.4 million barrels per day to 42.4 mb/d. The IEA cut its 2026 demand outlook to a contraction of 80,000 barrels per day on the year, an 810,000 barrels per day downward revision in a single month. Brent crude finished April more than 55% above pre-conflict levels, with North Sea Dated trading around $130 per barrel.² By any historical standard this is the setup for a major equity drawdown: the largest energy supply disruption in history, active U.S. military involvement in the Persian Gulf, a 2.3 million barrels per day demand contraction in April (the sharpest monthly drop since COVID), and a structurally fragile consumer. The S&P 500 closed April at an all-time high anyway.

D. The feedback loop inversion. The most insidious feature of the current regime is how it self-reinforces. The market habituates and does not sell off, Trump sees no selling and receives no feedback, and without feedback he escalates again, which the market then ignores. Each turn of that loop tightens the next one. The event that finally triggers true repricing will need to be structurally irreversible rather than merely loud, since loud is what the market has been trained to ignore, and by the time it reacts the policy mistake will already be embedded in the real economy.

Intellectual honesty requires engaging the strongest version of the counterargument, and habituation is not mass delusion. The market's recovery from Liberation Day lows was grounded in fundamental data, and the rally that followed has had real earnings support every step of the way. The legitimate version of the bull case is that the market has learned to filter signal from noise correctly, and that what looks like indifference is in fact discipline.

Earnings growth in 2025 was real. S&P 500 companies delivered double-digit annual EPS growth, and Q1 2026 is tracking 27.7%, the highest since Q4 2021. Net profit margins reached a record 13.4%.⁴ The AI capex cycle, with hyperscaler spending revised up 12% to roughly $720 billion for 2026, provides a multi-year demand pipeline for semiconductors, networking, power, and data center construction. Goldman Sachs projects roughly 12% index returns for 2026, anchored by megacap earnings power.¹¹ The One Big Beautiful Bill Act's R&D expensing provision likely contributed as a tax tailwind alongside the AI infrastructure cycle, with S&P 500 capex growing 39% year-over-year in Q1 2026 against just 1% growth in gross buybacks.¹¹

The bull case also rests on a distinction this report endorses. There is a difference between a market that is wrongly priced and a market whose signal function is impaired. The S&P 500 at 21x forward earnings may be approximately fair given the current trajectory of earnings, which means the problem at hand is not necessarily one of valuation. The problem is that the market has stopped transmitting risk information to policymakers, and an impaired signal function is a problem even in a market that is otherwise fairly priced. Jamie Dimon framed the point well during the April 2025 episode²²:

The productive framing is therefore not that the market is wrong, but that the market is calm for the wrong reasons. A patient whose pain receptors have been numbed has not healed, and a market that has stopped flinching has not stopped accumulating risk.

Markets habituate until they do not. The relevant question is not whether the spell breaks, since it will, but what is large enough to break through and how late the warning comes. Five conditions, any one of which would likely force true repricing, are worth tracking.

1. A VIX that stays elevated. A single-day spike is priced and forgotten, as the April 2025 episode demonstrated (VIX from a 60.13 intraday peak back below 20 in under 100 days).⁵ A VIX that stays elevated for weeks rather than collapsing back within days would indicate the market has materially repriced its distribution of outcomes, rather than merely absorbing a single piece of bad news.

2. Investment-grade credit spreads that keep widening. Investment-grade spreads sit well below their long-term average today.²³ A sustained, broad-based widening, rather than a brief blip that snaps back, would signal that bond markets are pricing default risk that equities cannot ignore.

3. Sustained, worsening dollar weakness. The dollar fell sharply through 2025.¹⁸ A further sustained leg lower, rather than a pause and stabilization, would signal not just trade policy disagreement but the beginning of structural diversification away from U.S. assets by foreign Treasury holders, a longer-duration problem that does not reverse with a Truth Social post.

4. Consumer spending contraction in hard data. Sentiment has already collapsed to record lows, which is soft data.¹² If that translates into negative retail sales prints, declining credit card volumes, or rising delinquencies, the fundamental override justification for equity resilience dissolves, since earnings depend on the consumer continuing to spend the way the top decile spends.

5. A policy shock without a reversal mechanism. Every previous shock contained an off-ramp: a court ruling, a presidential climb-down, an export carve-out. A debt ceiling crisis that reaches technical default, a Fed-independence confrontation that breaks the central bank's credibility, or a kinetic conflict that prevents the kind of de-escalation seen post-Hormuz, would represent a qualitatively different category of event, one that does not end with a Truth Social post.

The risk in the current setup is one of timing. The spell will break; the question is whether it breaks early enough to allow positioning, or late enough that the eventual move is larger and faster than anything 2025 participants had to manage.

The Numb Market is not a market that has conquered volatility but a market habituated to a sixteen-month cry-wolf policy cycle, until true risk signals have become indistinguishable from noise. The calm reads as quiet only because the alarm has been disconnected from the wiring.

The evidence has compounded. The Supreme Court voided most of the legal basis for the tariff regime in February 2026 and the market closed up 0.69%. The IEA called the Strait of Hormuz disruption the largest in oil supply history and the S&P 500 hit an all-time high. The University of Michigan Current Conditions index posted the lowest reading in the survey's history while forward P/E expanded to 21x. Each of these would historically have triggered material repricing on its own. Together, they have moved markets less than a typical earnings miss.

The two findings of this report reinforce each other in an uncomfortable direction. As markets desensitize, the cluster of signals that has historically constrained Trump loosens, which removes the primary mechanism by which markets have moderated erratic policy behavior; as that constraint weakens, policy escalation becomes more likely, and the structural risks (fiscal imbalances, consumer bifurcation, dollar depreciation, tariff permanence, kinetic conflict in the Persian Gulf) accumulate beneath the calm surface.

The market has stopped doing one of the two jobs it was supposed to do. Price discovery continues to function: earnings revisions and capex cycles move the index in a recognizable way, and the algorithms have not forgotten how to set spreads on Q2 2026 guidance. The other function, transmitting risk back to the policymaker who depends on the signal as a corrective, has gone offline. The bond market did that job in April 2025. The equity market has not done it since, even with a Supreme Court ruling that voided most of the legal foundation of the Liberation Day regime and an oil supply disruption the IEA calls the largest in history. A warning system that has gone quiet because it has stopped listening is exactly the system that fails when the wolf finally arrives.