The Bull Has a Breadth Problem. UBS Sees Upside Anyway

On September 18 — the day UBS told clients to keep pushing global equities higher, on resilient earnings and structural growth, over the next six to twelve months — only 49.5% of S&P 500 stocks traded above their 200-day moving average. Barely half. The index the bank is cheerleading, meanwhile, closed about 2.3% below its 52-week high (SPY), with the global proxy ACWI about 2.2% off its own.

On its face, the market UBS is describing looks undamaged. SPY is up 11.7% year to date and roughly +20% over the last 120 days; ACWI is up 12.6% YTD and about +18% over the same window. Looking under the hood is where the plot thickens.

Breadth has been quietly dismantled. In mid-August, about 72% of S&P 500 names sat above their 200-day line. By September 18 that figure had fallen to 49.5% — a twenty-plus-point fade in about a month, while the index gave up only a couple percent from its high. The 50-day reading ran even weaker: just 42% of stocks above it as of September 12. Put differently, the index is within reach of a record while the median member is below its trend line. A cap-weighted average only needs a few large names to keep printing highs; the rest of the board can be sitting in their own quiet drawdowns and the index won't show it. When the median stock slips under its trend line while the index dances near a record, the rally has stopped behaving like a tide and started behaving like a handful of boats.

UBS's fundamental case deserves stating cleanly before we pick at its internals. The bank pencils 26% earnings growth for the MSCI ACWI this year and 14% next, and expects AI capital spending to rise 84% this year — a real engine, not a hope. It prefers the eurozone, Asia ex-Japan and emerging markets within global equities, and figures a shallow rate-hike cycle won't derail the constructive view.

But the index that carries that earnings story has narrowed into a concentration record. The top ten names in the S&P 500 accounted for a record 41.2% of the index in December 2025 — a level that has since been exceeded as concentration has continued to climb — which means the index number is a slice of the market, not its census.

History, though, suggests the stretch itself isn't the tell. Returns after new all-time highs average higher than after other days; dumping everything because the index made a record has usually been the wrong move, as of this writing.

So the two readings score differently, and they don't have to agree. The fundamental case for global earnings is strong; the technical state of the median U.S. stock is weak. If UBS is right, the index keeps climbing on the strength of its leaders — the same AI-capex names that carried it here — while the half of the market already beneath its trend line has to earn its own way back. Which half a reader is actually buying is the question the breadth number begs. History only tells us not to panic-sell the record; it doesn't say which side carries the next leg. The number to watch is the one we opened with: if the 200-day breadth reading keeps sliding below half, the index and the median stock will eventually have to agree on which way the tape leans.