Base case: the S&P 500 reaches 7,650 by year-end 2026 on a 22.5× multiple applied to $340 of 2026 earnings per share. Bull case 8,100 (23.8×), bear case 7,100 (20.9×). The mechanism is multiple arithmetic, not earnings — Wall Street’s 1,150-point target spread rests almost entirely on what investors will pay, because the earnings forecasts have already converged.

Wall Street’s year-end targets for the S&P 500 now run from 7,100 to 8,250 — a 1,150-point, 16% spread. What makes that dispersion unusual is that the earnings assumptions underneath it barely differ. FactSet puts calendar-2026 earnings growth at 24.1%, Goldman Sachs models $340 of earnings per share on 24% growth, and no major desk is arguing the profit cycle breaks this year. The entire disagreement is about the multiple. This piece walks the arithmetic, the two camps, and the levels that would settle it.

Methodology

Targets and earnings figures are taken from published desk research and FactSet aggregates as reported between June 26 and July 5, 2026. Index levels are official closes to July 17, 2026. Implied multiples are calculated by dividing each published year-end target by Goldman Sachs’ $340 calendar-2026 earnings-per-share estimate — a deliberate simplification, since desks run their own earnings numbers and some are implicitly discounting 2027 earnings rather than 2026. The purpose is to express every target in a common unit, not to claim each firm uses this exact denominator. Caveat: a single shared EPS denominator understates dispersion where a desk’s own EPS estimate is materially above or below $340.

Key levels

• Spot: 7,457.69 at the July 17, 2026 close, down 76.08 points (-1.01%) on the day — official close
• Implied multiple at spot: 21.9× on $340 CY2026 EPS — author calculation
• Base target: 7,650 by December 31, 2026 (22.5×)
• Bull target: 8,100 (23.8×) — Citigroup year-end target
• Bear target: 7,100 (20.9×) — Bank of America year-end target
• Resistance: 7,620.90, the 52-week high
• Invalidation: a sustained close below 7,140 (21.0×) or above 8,160 (24.0×)

The arithmetic nobody states plainly

At 7,457.69 the index already trades at 21.9× the $340 earnings figure. That matters because the widely-cited assumption is that the US market multiple stays “roughly flat at 21 times earnings”. If that were the operative base case, fair value would be near 7,140 — some 4.3% below where the index closed on July 17.

So the market is not waiting for multiple expansion. It has already taken it. Anyone quoting a flat-21× assumption while holding a target above 7,200 is running an internally inconsistent model.

What multiple do the bulls actually need? Goldman Sachs’ 8,000 target on its own $340 EPS estimate implies 23.5×. Citigroup’s 8,100 implies 23.8×. Ed Yardeni’s 8,250 implies 24.3×. None of those is an earnings call — each is a bet that investors pay two to three turns more for the same profits than they do today. That is a legitimate view; equity multiples expanded through the late-1990s and 2020-21 in exactly this way. But it should be labelled accurately. The bull case for US equities in the second half of 2026 is a multiple-expansion thesis wearing the language of an earnings thesis, and the distinction determines which data actually falsifies it. Earnings beats will not validate a multiple call; falling real yields and sustained inflows will.

Where the targets sit in multiple terms

Firm / strategist	Year-end target	Implied multiple on $340 EPS	Upside from 7,457.69
Ed Yardeni (Yardeni Research)	8,250	24.3×	+10.6%
Citigroup	8,100	23.8×	+8.6%
Goldman Sachs	8,000	23.5×	+7.3%
JPMorgan	7,800	22.9×	+4.6%
This call (base)	7,650	22.5×	+2.6%
Bank of America	7,100	20.9×	-4.8%

Sources: published year-end targets as reported by Fortune (July 5, 2026) and Goldman Sachs research; EPS denominator $340 per Goldman Sachs; index level official close July 17, 2026. Implied multiples are author calculations.

The bull case, stated fairly

The earnings momentum is real and it is broad. FactSet’s blended second-quarter earnings growth rate stands at 23.3% year on year, with full-year 2026 growth at 24.1% (Forbes, July 5, 2026). A market compounding profits at that rate does not usually de-rate.

“The current bull market is driven by FEMO (fabulous earnings momentum),” said Ed Yardeni, President of Yardeni Research, in commentary published on July 5, 2026 (Fortune). Yardeni raised his year-end target to 8,250 from 7,700 in May.

Goldman Sachs makes the same argument structurally, projecting that earnings growth rather than re-rating powers the index higher (Goldman Sachs). The difficulty, per the table above, is that its own target requires 23.5× — the earnings do the work only if the multiple also cooperates.

The bear case, stated fairly

Bank of America holds the lowest published target at 7,100, with equity strategist Savita Subramanian warning that US stocks are flashing “too many red flags”. The firm’s research framed the condition bluntly: “Our bear market signposts suggest speculation is hitting extreme levels as high multiple stocks have gapped up demonstrably” (Fortune).

JPMorgan, despite raising its target to 7,800, cautioned that “the path higher is likely to be non-linear given a tougher bar into 2Q earnings, crowded Momentum positioning”. Crowding is the operative word: a multiple sustained by concentrated positioning unwinds faster than one sustained by flows.

The rate backdrop is the harder problem. Treasury yields have pushed to multi-year highs, and Kevin Warsh, who became Federal Reserve chair in May 2026, has brought a hawkish posture and scepticism toward the size of the Fed’s balance sheet. Multiples are a discount-rate function; a chair disinclined to cut, running a smaller balance sheet, is the single most direct threat to 23×. That dynamic is already visible in the currency, as our read on DXY’s path to 105 on Warsh repricing set out.

Why 22.5× is the base case

The index closed July 17 below its 50-day moving average, having fallen more than 1.5% on the week, led by semiconductors. That is a de-rating impulse, not an earnings event — profit estimates did not move.

Splitting the difference is not the reasoning. The reasoning is that 21.9× is where the market has settled with the profit cycle running near 24% and a hawkish Fed chair in place. Modest further expansion to 22.5× is consistent with earnings delivery continuing while the discount rate stays hostile. Full expansion to 23.5–24.3× requires the rate picture to soften, and nothing in the Warsh posture points that way this year. Our earlier Q3 multiple-compression call at 7,700 reached a similar destination by a shorter route; the year-end horizon simply gives earnings two more quarters to carry the level.

What would invalidate this call

• A sustained close below 7,140 (21.0×). That would confirm the flat-multiple assumption is binding and hand the argument to Bank of America’s 7,100.
• A sustained close above 8,160 (24.0×). Trading through the Yardeni multiple would mean the market is discounting 2027 earnings, and this framework’s denominator is wrong.
• CY2026 earnings growth revised below 20%. The whole structure assumes the profit cycle is not in dispute. If FactSet’s 24.1% aggregate falls under 20%, the multiple debate becomes secondary.
• An explicit Warsh signal on cuts. A Federal Open Market Committee (FOMC) statement opening the door to easing would compress real yields and make 23.5× reachable without any earnings change.

What to watch

Second-quarter reporting is the near-term catalyst, but the informative variable is not the beat rate — at 23.3% blended growth, beats are already assumed. Watch whether beats are rewarded. In a market at 21.9×, an in-line quarter that sells off signals the multiple, not the profits, is the binding constraint. Semiconductor leadership is the second tell, given it drove the July 17 decline. For cross-asset confirmation, our gold call to $4,500 tracks the same real-yield variable from the opposite side.

FAQ

Where did the S&P 500 close on July 17, 2026?
At 7,457.69, down 76.08 points or 1.01% on the day, and more than 1.5% lower on the week, with the index closing below its 50-day moving average.

What are Wall Street’s year-end 2026 targets?
They range from Bank of America at 7,100 to Ed Yardeni at 8,250, with Goldman Sachs at 8,000, Citigroup at 8,100 and JPMorgan at 7,800.

Why is the target dispersion a multiple argument rather than an earnings one?
Because the earnings assumptions have converged. FactSet puts calendar-2026 growth at 24.1% and Goldman models $340 of EPS; no major desk forecasts the profit cycle breaking. Expressed against that common denominator, the targets differ only in what investors will pay per unit of earnings — 20.9× at the low end, 24.3× at the high end.

What multiple does the index trade at now?
Roughly 21.9× on $340 of calendar-2026 earnings per share, which is already above the frequently-cited “flat at 21 times” assumption.

What is the single biggest risk to the bull case?
The discount rate. Treasury yields sit at multi-year highs and Fed chair Kevin Warsh, in office since May 2026, has signalled scepticism about the balance sheet and shown no easing bias. Multiples expand when real yields fall, and that has not happened.

This article is informational analysis only and is not financial, investment, or trading advice. Foreign-exchange, commodity, and equity markets are highly volatile and can lose substantial value rapidly. Leveraged products carry total-loss risk and may exceed the initial margin posted. Past performance and historical correlations do not guarantee future results. Do your own research and consult a regulated financial adviser before making any investment decision.