J.P. Morgan pegs the probability of a U.S. recession at 35%. Goldman Sachs says 30%. Those numbers have dominated headlines for a week. But flip them over: the same firms are telling you there is a 65 to 70% chance the expansion continues. That is the number I keep coming back to, because it aligns with what the actual earnings data shows. The recession-inevitable crowd is confusing elevated risk with a foregone conclusion, and the distinction matters for anyone making decisions with real money.
The Earnings Tell a Different Story Than the Headlines
S&P 500 earnings growth for 2026 is estimated above 16%, with Q4 2025 revenue and profits beating expectations. That is not a number you see on the eve of a contraction. In the 4 quarters before the 2008 recession began, S&P earnings were already declining year over year. Before the 2020 collapse, earnings growth had flatlined near zero. Right now, corporate America is printing money faster than it did a year ago.
Real consumer spending is projected to grow 2.1% this year. Slower than 2024, yes. But positive real growth means households are still buying more stuff after adjusting for inflation. A family earning $75,000 with 2.1% real spending growth is putting roughly $130 more per month into the economy than they were a year ago. That is not a consumer base in retreat.
Tax refunds running ahead of 2025 levels add a cash buffer that most recession models ignore. Lower borrowing costs from the Fed's 2024 and 2025 rate cuts mean a homeowner who refinanced a $350,000 mortgage from 7.5% to 6.5% is saving about $240 a month. Multiply that across millions of households and you have a quiet stabilizer that never makes the front page.
The Iran Premium Is Real but Priced
I will grant the bears their strongest card. The Strait of Hormuz carries 20 million barrels a day, and any prolonged disruption sends energy costs into a place that hurts everyone, especially lower-income households already squeezed by 4.1% CPI. Economist Saidel-Baker is right that the consumer base is bifurcated. The bottom quartile is stretched thin.
But the S&P 500 erased all Iran-related losses by early April. Markets are not ignoring the conflict; they priced it and moved on. The IMF's severe scenario of 2% global growth requires a prolonged Hormuz shutdown, and even the IMF calls that an adverse case, not a base case. Larry Fink's $150-per-barrel oil warning is a tail risk, not a forecast. Treating tail risks as certainties is how you miss the 65% of outcomes where the expansion holds.
The Q4 2025 consumption downshift that J.P. Morgan flagged is worth watching. It was real. But J.P. Morgan itself expects a rebound in the first half of 2026, driven by fiscal stimulus. When the same firm that identifies the slowdown also projects the recovery, the signal is deceleration, not collapse.
Vishal Garg of Better.com says he "personally believes a recession is coming," citing AI-driven layoffs. I respect the honesty of putting a name on a prediction. But personal belief is not an earnings report. AI workforce disruption is a structural story that plays out over 5 to 10 years, not a cyclical trigger for the next 12 months. Confusing the two leads to bad timing.
The right move for investors and households is straightforward: do not make defensive decisions based on a 35% probability. That means do not panic-sell equities into 16% earnings growth. Do not delay a home purchase because Goldman's model ticked up a few points. If you are in the lower-income bracket getting squeezed by food and energy costs, the policy answer is targeted relief, not a declaration that the whole economy is failing.
A 35% recession probability means the economy is under stress. It also means the base case is survival. I will take the base case backed by audited earnings over a headline backed by anxiety.
