WEEKLY MACRO UPDATE — August 16, 2026
─────────────────
MARKETS
S&P 500
Close: 7,785 | Weekly: +0.44%
CNN Fear & Greed: 65 — Greed
Nasdaq 100
Close: 30,046 | Weekly: +1.13%
Russell 2000
Close: 3,066 | Weekly: +1.16%
Dow Jones
Close: 53,680 | Weekly: +0.45%
RSP (Equal Weight S&P 500)
Close: 776 | Weekly: +0.48%
A quieter week on the surface, but structurally important.
The
S&P
took out its August 5 high on August 13, closed above it, and printed a fresh all-time high with a bullish weekly close, reaching 7,815 intraweek.
The
Nasdaq
did the same, taking out its August 5 high on the 13th and closing above it near 30,179, then retested that old high on Friday. On a daily basis, the Nasdaq also broke the trendline that had held since June 3, with a retest following, next week will be important for confirmation.
The
Russell
took out its August 5 high (3,050) on August 12 at 3,053, then built consecutive new highs Thursday and Friday, closing the week near 3,070.
RSP
behaved similarly to the S&P, printing a new high with a Friday retest of the August 5 level. Breadth confirmed the move alongside price.
─────────────────
CRYPTO
Bitcoin
: ~63,000 | Weekly: -2.94%
Crypto Fear & Greed: 37.4 — Fear
Ethereum
: 1,880 | Weekly: -1.87%
Bitcoin dropped from a local high near 65,400-65,500 on Saturday, August 9, to a low near 62,500 this week.
The bearish camp, particularly the calendar-model side, feels validated, the thesis that crypto tends to weaken in the second half of August before a bottom forms over the following 50 to 60 days. The market currently looks like it's searching for a catalyst.
Sentiment
remains unchanged from what we've written repeatedly: high consensus around either continued downside now, or a relief rally followed by the drop, largely in line with prior cycle behavior. Nothing here materially changes that view.
Worth noting
Michael Saylor's
recent modest selling briefly rattled the space, even though the size was small, itself a signal of how fragile current sentiment is.
─────────────────
DOTCOM FRACTAL — STATUS
All three fractals remain intact and continued advancing this week. The S&P, Nasdaq, and Russell all pushed to new highs, with the equal-weight index (outside the core fractal set but tracked as a breadth confirmation) mirroring the S&P's action closely. Full fractal update follows separately.
─────────────────
SECTOR ROTATION
Weekly performance:
XLE +6.39%,
KRE +2.65%,
XLU +1.94%,
XLP +1.85%,
IGV +1.82%,
XLF +1.99%,
XLRE +1.21%,
XLK +1.04%,
XLV +1.07%,
XLC +1.57%,
XLI +0.34%,
SOX +0.48%,
XLY -0.85%,
XLB -0.37%.
Key ratios:
IWM/SPY +1.1%,
XLK/XLU -1.03%,
XLY/XLP -2.5%,
RSP/SPY +0.6%.
A rotation week in the truest sense. Energy led decisively alongside oil's 5% weekly gain, while defensives (XLU, XLP) also outperformed even as the broader market pushed higher, an unusual combination. Consumer discretionary lagged staples by 2.5 points, and tech underperformed utilities, a reversal from the growth-led leadership of recent weeks. Small caps outpaced large caps and the equal-weight index outpaced cap-weighted, both signs of genuine broadening rather than narrow, mega-cap-only strength.
Full breakdown in this week's Sector Risk Monitor.
─────────────────
RATES, DOLLAR AND COMMODITIES
30-year: 5.26% | Weekly: +0.82%
10-year: 4.692% | Weekly: +0.60%
2-year: 4.171% | Weekly: -0.88%
Oil (WTI): $82.93 | Weekly: +5.06%
Gold: 4,376 | Weekly: +0.81%
Silver: 64.708 | Weekly: +1.74%
DXY: 99.636 | Weekly: +0.05%
The yield curve moved in two directions at once this week, a divergence worth watching closely. The 2-year fell nearly a full point while the 10-year and 30-year both rose. The short end tends to move first and hardest ahead of a slowdown, since it reflects near-term rate expectations more directly than longer maturities.
Oil's sharp weekly gain came alongside the Hormuz situation intensifying, discussed below.
─────────────────
GEOPOLITICS
President
Trump
said Friday he plans to declare the Strait of Hormuz a US territory once the conflict with Iran concludes, remarks made during a campaign event in New York. He also told the crowd that some near-term pain at the pump would be worth keeping the blockade in place.
Iran's government rejected the claim outright, stating Hormuz remains blocked regardless of who claims to control it. The comments landed as ceasefire negotiations remain stalled and a Quinnipiac poll from late July found 60% of US voters oppose continued military action.
Ukraine's campaign
against Russian oil infrastructure continued unabated, with the Slavneft-YANOS refinery in Yaroslavl, one of Russia's top-five facilities, struck for the seventh time this year. Ukrainian officials now put total 2026 refinery, terminal, and depot strikes at over 160 facilities, with roughly 37% of Russian refining capacity currently offline and refining output at a 24-year low according to Bloomberg.
─────────────────
MACRO
Consumer sentiment
delivered another data point in the same direction as everything else this week. The University of Michigan's preliminary August reading fell to 51.0, well below the 54.5 consensus and down sharply from July's final 55.2, ending a brief two-month recovery. The expectations component fell harder than current conditions, and only 8% of consumers now expect income growth to outpace inflation over the next year. Some argue these surveys carry political distortion and limited standalone signal, and there's a fair point in that critique.
But set against everything else in this update, the reading holds weight: it currently sits below where sentiment stood at the start of every recession since this survey began in 1978.
Retail sales
for July fell 0.6% against expectations for a 0.1% gain, roughly minus 0.7% to minus 0.8% in real terms. The weakness was broad, not isolated to one category: autos fell 1.8%, online retail fell 2.2%, and the GDP-relevant control group came in negative around minus 0.4%. The few categories that gained, restaurants, apparel, general merchandise, overlapped almost entirely with the World Cup running through July 19, a plausible temporary tailwind rather than organic strength.
Housing
data reinforced the same picture. Existing home sales fell 1.7% in July to an annualized 4.05 million units, even as median prices hit a record for the month at $434,100. More telling: Redfin's latest report shows the gap between buyers and sellers at the US housing market widening further, with sellers now outnumbering buyers by a record margin in the majority of markets tracked, and buyer demand at a record low.
─────────────────
EARNINGS REVIEW
The pattern that's defined this AI earnings cycle extended into names beyond the usual hyperscalers.
Applied Materials
beat on revenue and EPS, posted record numbers across the board, and raised guidance, yet fell over 5% after hours. The real issue: free cash flow collapsed 80% year over year to $210 million despite record net income, the fourth miss on cash flow in five quarters, as rapid shipment growth consumes cash faster than it generates it.
Cisco
delivered record revenue with beats on both lines and stronger-than-expected guidance, and still fell 7-9%, on gross margin compression and a slightly light Q1 outlook.
Cerebra
beat its own margin guidance but missed on revenue and fell 14%.
On the other side,
CoreWeave, Nebius, and Supermicro
all posted strong results and were rewarded, Nebius jumping 34% in a single session. Across the broader Q2 season, 86% of S&P 500 companies beat estimates with 51% aggregate earnings growth, healthcare remains the lone sector with declining profits.
─────────────────
EARNINGS AND MACRO — WEEK AHEAD
Home Depot and Walmart
both report next week, useful directional reads on the consumer even though Walmart's scale and business mix make it a less precise sentiment gauge than smaller retailers.
Target and TJX are also expected in the window. The bigger catalyst sits the week after: Nvidia
reports in late August, alongside the Fed's Jackson Hole symposium, both likely to set tone heading into September.
─────────────────
BOTTOM LINE
Markets right now are effectively split into two camps.
One expects a pullback first. The other
believes the next multi-year cycle in stocks, and in crypto, has already begun. What neither camp fully reconciles with is the consumer, because the data supports neither story cleanly. It doesn't look like a normal seasonal correction that resolves and resumes. And it doesn't look like the start of a multi-year expansion either. Consumer sentiment is the logical output of that gap, not a standalone anomaly.
The 2026 monthly average for
nonfarm payrolls
, after all revisions, sits at roughly 69,000 jobs per month. Compare that to 1999, which averaged 229,000 per month for the full year, or 2007, which averaged around 111,000, already a sharp slowdown heading into the financial crisis. Inflation cooling further actually makes sense in this context: a weakening labor market and a softening housing market both apply deflationary pressure, not inflationary. Friday's retail sales miss fits the same pattern, and likely would have looked worse without the World Cup tailwind running through most of July.
Put together: nonfarm payrolls at historic lows outside of recession, a widening gap between the official U-3 unemployment rate and the broader U-6 measure, retail sales contracting even with a tailwind, and now consumer sentiment sitting below where it stood at the start of every recession since 1978. This is what early-stage weakness in the system looks like.
Against that backdrop, what we have is a market that largely distrusts its own rally, leans on fractals from prior years, and concludes a larger drawdown is coming. That's always possible.
But parabolas operate under their own rules and carry their own signatures. Sharp, fast pullbacks that resolve structural excess quickly, without breaking the larger trend, are entirely consistent with a parabola continuing higher afterward. Momentum still holds as long as there's no structural break. For the S&P, that break would mean falling back into the June-July correction zone and taking out the June 9 low. The Nasdaq's equivalent level functions the same way.
While the macro backdrop increasingly fails to support the consumer strength an expanding economy and rising markets require, and while this rotation market keeps building parabolic structures that shake out leveraged positioning and detach sentiment from fundamentals along the way, crypto continues to exist in its own bubble, one where the bear market is expected to wrap up quietly over the next 50 to 60 days before the next cycle begins on schedule.
For most crypto analysts, crypto sits at the center of the analysis, and everything else gets fit into that frame however convenient, rather than the other way around. Since most people's interest in markets began through crypto itself, the financial world effectively starts around 2008 for many of them, and data from before that period mostly gets used selectively, to reinforce the existing model.
Based on the current data, that model is worth doubting. Something different could easily unfold over the coming weeks and months, and the idea of a simultaneous multi-year bull cycle in equities and crypto runs into the same problem either way: the consumer situation, and the labor market specifically, look nothing like they did heading into either the Dotcom top or the run-up to the financial crisis.
