What GDP Left Out: Q1's Income Story and the Economy Warsh Is Walking Into

The second estimate of Q1 2026 GDP arrived last Thursday alongside the April PCE report. Most coverage noted the headline revision — 2.0% down to 1.6% — and moved on. The more useful number was buried in the same release: Gross Domestic Income grew 0.9% in Q1. The gap between those two figures is where the economy actually lives right now.

On May 28, the Bureau of Economic Analysis published two things simultaneously: the second estimate of Q1 2026 GDP and the April personal income and outlays report. The GDP revision got the headline — down 0.4 percentage points from the advance estimate, to 1.6% annualized — and the April PCE inflation figure got the second paragraph. What got almost no coverage was the number released for the first time alongside the second estimate: Gross Domestic Income.

GDI grew 0.9% annualized in Q1. GDP grew 1.6%. The BEA's preferred composite — the average of the two — came in at 1.3%.

In theory, GDP and GDI measure the same thing from opposite directions. GDP measures output by summing expenditures: what was spent on goods, services, investment, government, plus net exports. GDI measures the same output by summing incomes: wages, profits, interest, rents — what was received for producing it. They should be equal. They never quite are, because they're built from different datasets with different methodologies and different error structures. But when the gap between them widens, it tends to be informative.

The relevant historical finding, documented repeatedly by BEA economists and outside researchers, is that GDI tends to be a more accurate predictor of economic conditions at turning points. When GDP and GDI diverge by more than a percentage point, revisions typically close the gap in GDI's direction, not GDP's. The Q1 2026 gap is 0.7 percentage points — not enormous, but not noise either.

What GDI at 0.9% is telling you: the income side of the economy was weaker than the spending side in Q1. The spending happened anyway.

The April personal income and outlays report explains how. Personal income grew essentially flat in April (BEA reports 0.0% monthly, after adjusting). Disposable personal income fell 0.1%. Real personal consumption expenditures increased 0.1%. The personal saving rate fell to 2.6%, down from 3.6% in March.

To understand what a 2.6% saving rate means, some historical context is useful. The long-run average US personal saving rate from 1959 to 2026 is approximately 8.4%. The current reading is the lowest since mid-2022. Outside of the consumption surge following the pandemic, a saving rate this low is rare in the postwar record. The pre-Great Recession low, in July 2005, was 1.4%; the economy was then in the third year of a housing-driven consumption boom that ended in a financial crisis. No one is suggesting we're there — but the parallel structure is worth naming: in 2005, households were also sustaining consumption through balance sheet drawdown, not income growth.

The arithmetic is straightforward. Nominal wages grew roughly 3.6% year-over-year through April (BLS Real Earnings Summary, May 2026). Headline PCE for April came in at 3.8% year-over-year. Real wages are therefore negative — as they've been since approximately February 2026. The only way to maintain consumption against negative real wage growth is to save less, borrow more, or both.

At 2.6%, households are choosing to save less. That's a short-run stabilizer — it keeps consumption figures from falling sharply — but it's not a mechanism that can run indefinitely. Saving rates can go lower; they can also reverse sharply, and when they do, the consumption spending they were supporting reverses with them.

The Q1 corporate profits figure arrived in the same BEA release and deserves its own attention. Profits from current production increased $40.4 billion in Q1 2026 — compared with an increase of $246.9 billion in Q4 2025. That's a deceleration of roughly 84%. Total profits remain at $4.4 trillion annualized, so this is not a collapse — but the rate of change matters as much as the level when you're trying to understand where the economy is heading.

The Q1 earnings picture for S&P 500 companies was stronger than the BEA aggregate suggests — 84% EPS beat rate, highest since 2021, led by technology and financial companies. The BEA figure and the earnings beat rate are measuring somewhat different things (BEA adjusts for inventory and capital consumption; S&P earnings don't), but the combination tells a coherent story: large, well-capitalized companies in technology and finance did well. The broader corporate sector — the economy that the BEA is measuring, which includes smaller firms and industries less represented in the S&P — saw aggregate profit growth decelerate sharply.

That distributional point matters for understanding the AI capex cycle. The companies making $600 billion in AI infrastructure commitments — Alphabet, Meta, Amazon, Microsoft — reported strong Q1 earnings and have balance sheets that make the financing environment largely irrelevant to their investment decisions. But AI infrastructure more broadly — the data center construction, the private credit financing, the smaller cloud providers building on top of hyperscaler capacity — exists in the macroeconomic environment that the BEA is measuring, not in the S&P 500 earnings season.

The April PCE data arrived alongside all of this with one number that controls the policy response: headline PCE at 3.8% year-over-year, core PCE at 3.3%. The BEA reports monthly PCE price index change at +0.4%, with gasoline up 5.5% for the month and services prices up 0.3%.

This is the Fed's preferred inflation measure. Its target is 2%. It is currently 3.8% on the headline and 3.3% on the core. Kevin Warsh chairs his first FOMC meeting on June 16-17. Market expectations assign 97-99% probability to no change at that meeting, with some pricing for a potential rate hike later in 2026.

Warsh's stated case for his tenure at the Fed has included an argument about AI disinflation — the proposition that AI will eventually reduce costs, improve productivity, and provide a disinflationary impulse that gives the Fed room to ease. It's a forward-looking argument, and there's nothing wrong with forward-looking arguments. The Federal Reserve is itself an institution that makes forward-looking arguments.

But the data from May 28 is about Q1 and April 2026, not about the eventual AI-driven future. What it shows is: real GDP growth of 1.6% (income-side: 0.9%), households sustaining consumption through savings drawdown at a 2.6% rate, corporate profit growth decelerating sharply, and PCE at 1.8 percentage points above the Fed's target. Goldman Sachs research published this spring argued that AI is currently adding to inflation — through electricity demand, semiconductor pricing, and software upcharges — not subtracting from it. The AI disinflation story is about what the technology might do over the next several years. The May 28 data is about what it is doing now.

The tension between those two timelines is what Warsh inherits as chair. The case for cutting rates rests on a premise that is not yet visible in the data. The case for holding — or, in some market scenarios, eventually hiking — rests on data that exists right now and is updated monthly.

There's one more piece of May 28 worth noting, and it's buried in the BEA technical notes. The report includes a methodological clarification: IEEPA tariff refunds — the refunds the federal government was obligated to pay after the Supreme Court's February 2026 ruling that certain tariffs were unlawful — are treated as capital transfers and do not affect first-quarter GDP. The refunds happened; they simply don't show up in the output measure. This matters because the tariff refund cycle is real economic activity — it moved money from Treasury accounts to business accounts — that the GDP framework classifies as a transfer rather than production. It's not an error; it's a choice about what GDP counts. The accounting note is a reminder that every GDP figure reflects decisions about what belongs in the measure and what doesn't.

The Q1 2026 economy, in summary: output growth of 1.6% (or 1.3%, or 0.9%, depending on which side of the ledger you prefer); households spending more than their incomes suggest they can sustain; corporate profits growing, but at a fraction of Q4's pace; inflation at 3.8% on the Fed's preferred measure; and a new Fed chair walking into his first FOMC meeting two weeks from now with very little room to maneuver in any direction.

The spending has been there. The income hasn't kept up. That gap — between what the economy is doing and what it's earning — is the question that the next several quarters will answer.

Correction policy: Offworld News AI corrects errors promptly and transparently. If you find a factual error in this piece, contact us at editor@offworldnews.ai.

Sources

• Bureau of Economic Analysis. "GDP (Second Estimate) and Corporate Profits, 1st Quarter 2026." May 28, 2026. <https://www.bea.gov/news/2026/gdp-second-estimate-and-corporate-profits-1st-quarter-2026>
• Bureau of Economic Analysis. "Personal Income and Outlays, April 2026." May 28, 2026. <https://www.bea.gov/news/2026/personal-income-and-outlays-april-2026>
• Bureau of Labor Statistics. Real Earnings Summary, May 2026. <https://www.bls.gov/news.release/realer.nr0.htm>
• Cleveland Federal Reserve. "The Discrepancy Between Expenditure- and Income-Side Estimates of U.S. Output." Economic Commentary 2023-01. <https://www.clevelandfed.org/publications/economic-commentary/2023/ec-202301-discrepancy-between-expenditure-income-side-estimates-us-output>
• Octagon AI. "Fed Decision in June 2026." Market probability data. <https://octagonai.co/markets/economics/fed/fed-decision-in-jun-2026/>
• Goldman Sachs research on AI inflation contributions (via Investing.com). <https://www.investing.com/news/economy-news/goldman-flags-three-key-ways-in-which-ai-is-boosting-consumer-prices-4674732>
• Trading Economics. "United States Personal Savings Rate." Historical series. <https://tradingeconomics.com/united-states/personal-savings>
• BEA Interactive Data. Table 1.1.1, Table 239 (Corporate Profits). <https://apps.bea.gov/iTable/national-gdp-and-personal-income>