Why the Dow Fell 1,153 Points After the Fed Held Rates Steady

Last updated: July 30, 2026, 5:15 a.m. EDT

The Dow Jones Industrial Average fell 1,153.18 points on July 29, 2026, even though the Federal Reserve did not raise interest rates. The apparent contradiction is the key to understanding the day. Markets were not reacting only to the policy rate that remained unchanged. They were repricing a more complicated combination of risks: stubborn inflation, a divided Federal Open Market Committee, higher long-term Treasury yields, a renewed oil shock, weakening confidence in the technology trade, and uncertainty about how Chair Kevin Warsh intends to communicate future policy.

The Fed kept its target range for the federal funds rate at 3.50% to 3.75% in a 9–3 vote. Three officials wanted a quarter-point increase. That was not a rate hike, but it was hardly a conventional dovish pause. Warsh emphasized that inflation remained too high, resisted giving markets a clear path for September, and defended a reduced reliance on forward guidance. Short-dated Treasury yields fell as traders marked down the probability of an immediate increase, while long-dated yields rose as investors demanded more compensation for inflation and policy uncertainty. That steepening of the yield curve was especially uncomfortable for richly valued stocks.

The selloff had already begun before the Fed announcement. Oil prices surged as fighting in the Middle East resumed and U.S. crude inventories fell sharply. Semiconductor shares were under pressure after disappointment around the Asian chip sector, and the Nasdaq 100 had already entered correction territory. The Fed press conference intensified an existing risk-off move rather than creating the entire decline from nothing. Microsoft and Meta then reported after the closing bell, producing sharply different-looking results that complicated the next day’s outlook rather than delivering a simple rescue for technology stocks.

Bitcoin, meanwhile, traded near $64,000 during the same period. Its relative stability compared with the equity selloff attracted attention, but a few hours of holding one price area did not establish that a major rally was imminent. The stronger conclusion is that July 29 offered a concentrated view of the forces now driving markets: the price of money, the credibility of inflation policy, the cost of energy, the durability of artificial-intelligence spending, and the willingness of investors to keep paying premium valuations when long-term yields are rising.

Key Takeaways

• Federal Reserve decision: The FOMC held the federal funds target range at 3.50% to 3.75% on July 29, 2026, by a 9–3 vote.
• Dissent: Beth Hammack, Neel Kashkari and Lorie Logan preferred a 25-basis-point increase, an unusually visible division early in Kevin Warsh’s chairmanship.
• Market close: The Dow lost 1,153.18 points, or 2.2%; the S&P 500 fell 1.5%; and the Nasdaq Composite declined 1.7%.
• Bond signal: The two-year Treasury yield fell while longer-term yields rose, with the 30-year yield briefly moving above 5.20%, indicating less conviction about an immediate hike but more concern about long-run inflation and policy risk.
• Oil shock: Brent crude rose about 8% to settle near $90.74 a barrel as Middle East hostilities resumed and U.S. inventories tightened.
• Technology earnings: Microsoft reported 18% revenue growth and 43% Azure growth, while Meta posted 28% revenue growth but lower profit, a narrower operating margin and only $784 million of quarterly free cash flow after heavy capital spending.
• Fact-check: The Fed’s new task forces are intended to review communication, balance-sheet policy, data, productivity and inflation frameworks. They are not an announced investigation into whether official economic statistics are “rigged.”
• What comes next: The July 30 GDP and inflation releases, August employment and inflation reports, oil prices, long-term Treasury yields and the September 15–16 FOMC meeting will shape the next policy decision.

What Happened on July 29, 2026

U.S. stocks opened under pressure and deteriorated through the session. By the close, the S&P 500 had fallen 112.63 points to 7,316.15, the Dow had declined 1,153.18 points to 51,594.14, the Nasdaq Composite had lost 433.97 points to 24,442.94, and the Russell 2000 had dropped 47.49 points to 2,906.31. Those percentage declines—1.5%, 2.2%, 1.7% and 1.6%, respectively—were more informative than the raw point totals because index levels are much higher than they were during earlier market eras.

The Dow’s four-digit loss was visually dramatic, but the day was not equivalent to a historic one-session crash. A 1,153-point decline from an index above 52,000 represents roughly 2.2%. The percentage move was still severe for a single session, particularly because it was broad enough to involve industrials, financial conditions and technology, but the point figure alone exaggerates comparisons with periods when the Dow traded at much lower levels. The correct description is a sharp selloff, not proof of systemic market failure.

Technology pressure was already visible before the Fed decision. Semiconductor shares had been sliding as investors questioned whether extraordinary AI-related capital spending would continue to translate into equally extraordinary earnings growth. Weakness in South Korean chip shares added to the concern, and the Philadelphia semiconductor index had suffered a deep decline from its recent peak. A market that had been rewarding nearly any company associated with computing infrastructure was becoming more selective about orders, margins, financing needs and the timing of returns.

Energy added a separate inflationary shock. Brent crude futures settled $6.65 higher at $90.74 a barrel, an increase of 7.91%, while West Texas Intermediate rose $5.20 to $84.46. The advance followed renewed airstrikes in the Middle East and a report showing U.S. crude inventories at a multi-year low. A sudden oil increase can hurt stocks through several channels at once: it raises costs for businesses, compresses household purchasing power, complicates the inflation outlook and reduces the Fed’s freedom to lower rates.

The market therefore arrived at the 2 p.m. EDT policy decision already carrying geopolitical, energy and technology risk. The Fed did not need to surprise investors with an actual increase to make conditions worse. It only needed to leave investors uncertain about whether inflation would require higher rates later, while doing little to relieve pressure at the long end of the Treasury market. That is what happened.

Source: Associated Press market close data for July 29, 2026. Index figures represent regular-session closes.

Why the Dow Fell After the Fed Decision Even Without a Rate Hike

The simplest explanation is that markets trade on the expected future path of financial conditions, not merely on the current policy rate. Holding the federal funds range steady prevented an immediate tightening, but the vote, statement and press conference all affected expectations for September and beyond. Investors had to assess not just whether the Fed had raised rates on July 29, but whether inflation might force it to maintain restrictive policy longer or resume increases later.

A central bank pause can be bullish when it arrives with falling inflation, easing wage pressure, stable energy prices and confidence that cuts are approaching. The same pause can be bearish when inflation is above target, oil is surging, long-term yields are rising and policymakers are divided. July 29 resembled the second configuration. The policy rate did not change, yet the risk around future rates and inflation appeared to increase.

The 9–3 vote mattered because dissents communicate the range of views inside the committee. Beth Hammack, Neel Kashkari and Lorie Logan preferred a 25-basis-point increase. Their votes did not bind the majority, but they showed that a meaningful bloc believed the existing setting might be insufficient. Investors could not treat the hold as evidence that the tightening debate was over.

Warsh also declined to offer the kind of direct, calendar-like guidance markets often seek. He acknowledged that policy rates “could well” be part of the solution if inflation remained elevated, while emphasizing that rates would not operate in isolation. The formulation kept both a future hike and a continued hold in play. That may be intellectually defensible, but it leaves traders with a wider distribution of outcomes—and uncertainty itself carries a price.

The stock market’s reaction also reflected the interaction between policy and valuation. Many large technology companies are priced on the assumption that substantial profits will arrive years into the future. When long-term Treasury yields rise, those future cash flows are discounted at a higher rate. Even if a company’s operating outlook is unchanged, the present value assigned to distant earnings can fall. The effect is especially pronounced for companies trading at high earnings multiples.

That valuation channel helps explain why the long end of the Treasury curve mattered more than the unchanged overnight rate. The Fed controls the federal funds target directly, but ten-year and thirty-year yields are set in markets and incorporate expectations for inflation, growth, federal borrowing, term premium and policy credibility. Investors were effectively saying that the absence of an immediate hike did not resolve longer-run concerns.

A Hawkish Hold, but Not in the Usual Sense

“Hawkish hold” is an imperfect but useful description. In central-bank language, a hawk prioritizes inflation control and is more willing to accept tighter financial conditions; a dove places relatively greater weight on employment and growth risks. The July decision combined an unchanged rate with an emphasis on persistent inflation and three votes for tighter policy. That mixture was more restrictive than a routine pause, even though Warsh avoided committing to the next move.

The Fed’s official statement said economic activity had continued to expand at a solid pace. It highlighted strong productivity growth and capital expenditure, particularly in high-technology sectors. Job gains had kept pace with labor-force growth, while unemployment was little changed. None of those observations described an economy urgently requiring stimulus. The statement’s concern remained inflation, which was elevated relative to the 2% objective and influenced by supply shocks and energy costs.

Warsh’s opening statement sharpened that message. He argued that more than five years of above-target inflation could not be declared over because of one favorable monthly report. That was a direct warning against extrapolating June’s softer consumer-price reading. It also signaled that the committee wanted a sequence of evidence, not a single data point, before concluding that inflation was returning sustainably to target.

At the same time, Warsh resisted the idea that the Fed must mechanically raise rates whenever employment is strong. Monetary policy works through many channels, and inflation can be driven by supply constraints, energy, fiscal conditions, productivity, expectations and the central bank’s balance sheet. A strong labor market does not automatically require a hike, just as a weak labor market does not automatically justify a cut. The relevant question is how all available information affects the outlook and the balance of risks.

This nuance was partly lost in same-day commentary. It is accurate to say that Warsh rejected a simple one-variable model in which jobs and inflation always move in opposite directions. It is not accurate to conclude that the Fed believes rates no longer influence demand or inflation. The committee still uses the policy rate as its primary tool. Warsh’s point was that the correct setting depends on the source and persistence of the inflation pressure, not on a single rule of thumb.

The Yield Curve Delivered the Day’s Most Important Signal

The most revealing market move occurred inside the Treasury curve. Two-year yields, which are highly sensitive to expectations for near-term Fed policy, declined after the decision. Longer-term yields rose, with the thirty-year yield briefly moving above 5.20%, its highest level since 2007. The ten-year yield also climbed, reaching roughly 4.67% during the session. This was not a uniform rise in rates; it was a steepening curve.

A steepening move of this kind can carry several interpretations. Traders may have become less convinced that the Fed would hike immediately, pushing down short yields. At the same time, investors may have demanded a higher term premium to hold long-dated bonds because of uncertainty about inflation, future government borrowing or the Fed’s ability to anchor expectations. The combination is uncomfortable: near-term policy may be less restrictive than feared, but longer-run financing conditions can still tighten.

Term premium is the extra return investors require for taking the risk of holding a long-duration bond rather than repeatedly rolling over short-term debt. It is not directly observable and must be estimated, but the concept helps explain why long yields can rise even when the expected near-term path of the federal funds rate falls. Inflation uncertainty, fiscal supply, volatility and reduced confidence in forward guidance can all increase that required compensation.

For equities, a higher thirty-year yield affects more than theoretical valuation models. It influences corporate borrowing costs, pension calculations, infrastructure finance, real-estate capitalization rates and the relative appeal of bonds versus stocks. A long Treasury yield above 5% gives investors a substantial nominal return from an asset with no corporate earnings risk. That raises the hurdle rate for paying elevated multiples on stocks.

The curve move also complicated the interpretation of the Fed’s decision. A central bank can hold its own rate steady while financial conditions tighten elsewhere. Mortgage rates, corporate yields and long-term project financing are tied more closely to Treasury benchmarks and credit spreads than to the overnight federal funds rate alone. Households and businesses may therefore experience tighter conditions even without a formal policy increase.

That is one reason Warsh’s reduced emphasis on forward guidance is consequential. If markets receive fewer signals about the central bank’s reaction function, they must assign greater probability to a wider range of future outcomes. Greater uncertainty can show up as volatility and term premium. The Fed may welcome a market that does not wait for every official speech, but independence from central-bank guidance does not guarantee lower borrowing costs.

Kevin Warsh’s Communication Experiment

Warsh became Federal Reserve chair in May 2026, returning to an institution where he previously served as a governor from 2006 to 2011. The July meeting was only his second as chair. That matters because markets are still learning how he frames trade-offs, how much weight he places on market prices, and how he distinguishes a deliberate absence of guidance from an absence of a policy framework.

His rhetorical style is notably different from the highly scripted language associated with many central bankers. During the July press conference, he used memorable phrases and engaged reporters directly. One line—“play the ball, not the referee”—captured his preference for investors to focus on economic information rather than trying to decode every Fed signal. The presentation may feel more conversational, but style does not eliminate the need for a predictable reaction function.

Forward guidance became especially important after the global financial crisis, when policy rates approached their lower bound and central banks tried to influence longer-term borrowing costs by describing the likely future path of policy. It can reduce uncertainty and make policy more powerful. It can also become a constraint when conditions change, encourage one-way market positions, or create the impression that forecasts are promises.

Warsh’s critique addresses real weaknesses. Markets can become overly dependent on “dot plots,” speeches and carefully parsed adjectives. An excessive focus on the Fed can amplify herding and weaken price discovery. If investors believe the central bank will always guide them away from losses, risk-taking can become distorted. Allowing rates to move in response to data rather than a steady stream of official hints can improve discipline.

The skeptical case is equally strong. Markets do not stop forming expectations when the Fed speaks less. They rely more heavily on partial data, political signals, individual policymakers and their own assumptions. A reduction in guidance can increase rather than reduce speculation. The Reuters Breakingviews analysis published before the meeting described the risk as a communication own goal: less guidance may produce more volatility without creating better understanding.

The July response suggests that investors heard Warsh’s confidence but did not receive a sufficiently clear map of what would trigger a hike, a hold or a cut. The result was not panic about one sentence. It was a repricing of uncertainty across the yield curve. Over time, the Fed can address this by making its reaction function clearer without promising a particular decision. It can explain which inflation measures, labor indicators and financial conditions would materially change the balance of risks.

What the Fed’s Five Task Forces Actually Do

One of the most important corrections to circulating commentary concerns the Federal Reserve’s task forces. The Fed announced five groups on July 9 to examine communications, balance-sheet policy, data, productivity and jobs, and inflation frameworks. Their purpose is to review how the institution operates and interprets information. The announcement did not say the Fed had created an investigative unit to determine whether government or private labor and inflation reports were manipulated.

The Data Task Force is intended to consider how to improve the quality and timeliness of real economic signals available to policymakers. That is a legitimate concern. Economic statistics are revised, surveys face declining response rates, new forms of activity can be difficult to measure, and real-time indicators may diverge from official releases. Improving data does not imply that existing numbers are fraudulent.

The Productivity and Jobs Task Force is focused partly on artificial intelligence and structural economic change. The Communications Task Force examines how the Fed explains policy. The Balance Sheet Policy Task Force addresses the role and composition of central-bank assets. The Inflation Frameworks Task Force considers how policymakers should think about price stability. Each group includes outside experts, including former central bankers, economists and technology figures.

Official statistics should be scrutinized. The Bureau of Labor Statistics publishes methodologies, revisions and survey limitations, and analysts regularly debate seasonal adjustment, sampling error and benchmark changes. ADP’s private payroll estimate uses a different data source and methodology from the BLS establishment survey. Disagreement between them is not proof that either series was engineered to influence rates.

The distinction matters because claims that data are “rigged” can undermine trust without evidence. A responsible analysis can question methodology, examine revisions and compare independent series. It should not convert uncertainty into an allegation of manipulation. The Fed’s initiative is best understood as an attempt to improve measurement and policy interpretation during a period of rapid economic change.

Inflation: One Good Month Was Never Enough

The June Consumer Price Index provided genuine relief but not a complete resolution. The all-items CPI fell 0.4% on a seasonally adjusted monthly basis and was 3.5% higher than a year earlier. Core CPI, which excludes food and energy, was unchanged for the month and increased 2.6% over twelve months. Those were materially softer monthly readings, especially for core inflation, but the annual headline rate remained far above the Fed’s 2% objective.

A slowdown in inflation is not the same as falling prices. When the annual inflation rate declines, the overall price level can continue to rise, just more slowly. A negative monthly headline reading can produce an outright one-month decline, often because of energy, but it does not reverse the cumulative increase in living costs over several years. Warsh’s emphasis on the five-year history addressed that distinction.

Energy was already distorting the data before the July 29 oil surge. In June, the BLS reported that energy prices were 15.7% higher than a year earlier and gasoline was up 26.7%. A fresh increase in crude prices does not pass through one-for-one or immediately to consumer inflation, but it raises the risk that transportation, production and utility costs will remain elevated. It can also affect inflation expectations, which matter because expectations influence wage bargaining, pricing and long-term yields.

Core inflation is usually less volatile and can offer a better view of underlying pressure, but policymakers cannot ignore energy when it changes household behavior and business costs. The Fed’s challenge is to distinguish a temporary geopolitical price shock from a broader process in which firms pass higher costs across the economy and workers seek compensation for lost purchasing power. The policy response to those two cases need not be the same.

The Personal Consumption Expenditures price index, the Fed’s preferred inflation measure, is broader than CPI and uses different weights. At the article’s research cutoff, the June PCE report had not yet been released; the Bureau of Economic Analysis had scheduled it for 8:30 a.m. EDT on July 30. In May, the PCE price index rose 0.4% for the month and core PCE rose 0.3%. The pending June data therefore represented an immediate test of whether the softer CPI signal would carry into the Fed’s preferred gauge.

Inflation policy also depends on expectations about persistence. A central bank may look through a temporary oil shock if demand is weakening and inflation expectations remain anchored. It may respond more forcefully if repeated shocks are being incorporated into wages, services prices and long-term contracts. Warsh’s refusal to commit before seeing more data reflects that uncertainty, though it also makes the market’s task harder.

The Labor Market Is Slower, Not Broken

June payroll employment increased by 57,000, and the unemployment rate was 4.2%. Those figures describe a labor market that had cooled from earlier strength but had not collapsed. Payroll gains were modest, participation was 61.5%, and the number of long-term unemployed remained elevated compared with a year earlier. The data support a cautious Fed, not an obvious one-direction policy response.

Employment matters for inflation because a tight labor market can support wage growth and household demand. Yet the relationship is not mechanical. Productivity can allow wages to rise without generating the same unit labor cost pressure. Labor supply can expand. Sectoral shortages can coexist with broader softness. Fiscal and energy shocks can raise prices even when hiring slows. That is why Warsh resisted treating jobs and inflation as a simple seesaw.

The central bank’s dual mandate requires maximum employment and stable prices. When inflation is above target and unemployment is stable, the Fed has more room to prioritize price stability. When the labor market weakens sharply, the cost of further tightening rises. July’s decision reflected a committee that saw enough growth to avoid easing but not enough evidence to support a majority hike.

The three dissents show how reasonable policymakers can read the same data differently. A hawkish interpretation emphasizes years of above-target inflation, strong activity and a new oil shock. A more patient interpretation emphasizes weaker payroll growth, uncertainty about the durability of energy inflation and the lagged effect of already-restrictive policy. The 9–3 vote placed the majority in the patient camp for July, but it did not close the debate.

AI Investment Is Both a Growth Story and a Policy Complication

Warsh devoted unusual attention to productivity and capital spending related to artificial intelligence. He said four-quarter growth in high-tech equipment and software investment was close to 20%. That surge can support future economic capacity by improving automation, data analysis, software development and decision-making. If output per worker rises, the economy can grow faster without generating the same inflation pressure.

The timing is the problem. Data centers require enormous quantities of chips, power, construction, cooling equipment and network capacity before the productivity benefits are fully realized. The investment boom can therefore raise demand for scarce resources in the near term while expanding supply only later. Monetary policy must decide whether the initial demand shock is inflationary, whether future productivity is credible, and how quickly the transition occurs.

Financial markets face the same timing mismatch. Companies are spending cash now for revenue and efficiency gains that may arrive over several years. Investors have tolerated that gap when cloud growth, advertising improvement or software adoption provide visible evidence of returns. They become less patient when capital expenditure rises faster than cash generation or when management cannot explain the path from infrastructure to profit.

Microsoft and Meta offer a useful comparison because both are spending aggressively, yet their June-quarter results presented different cash-flow profiles and narratives. Microsoft paired 43% Azure growth with record cloud revenue and a large increase in contractual backlog. Meta paired 28% revenue growth with a sharp rise in costs, legal and severance charges, a lower operating margin and free cash flow of only $784 million. The difference is not that one company “understands AI” and the other does not. It is that the market could more easily connect Microsoft’s spending to contracted cloud demand in this quarter.

That distinction helps explain the wider technology selloff. The market is moving from a phase in which AI capital spending was treated as inherently bullish to one in which investors ask who receives the spending, who monetizes it, who finances it, and how quickly the return appears. Rising long-term yields make those questions more urgent because the cost of waiting has increased.

Oil Turned a Fed Meeting Into a Stagflation Test

The July 29 oil move was large enough to alter the policy conversation. Brent’s 7.91% increase to $90.74 followed renewed Middle East airstrikes and a sharp draw in U.S. crude inventories. Oil had fallen the previous day on hopes for a pause in fighting, then reversed as hostilities resumed. That volatility demonstrated why the Fed cannot rely on a smooth forecast for energy prices.

An oil shock can create a stagflationary combination: higher inflation and weaker real growth. Households spend more on gasoline and utilities, leaving less for discretionary goods and services. Airlines, logistics companies, manufacturers and chemical producers face higher input costs. Some firms absorb the increase through lower margins; others pass it to customers. Either way, the shock can reduce economic welfare.

For the Fed, the difficult question is whether raising rates can solve a supply problem. Higher borrowing costs do not produce more oil. They can, however, prevent a temporary price shock from broadening into persistent inflation by cooling demand and anchoring expectations. The trade-off is that tightening into an energy shock can deepen the growth slowdown.

This is why the selloff cannot be attributed entirely to Warsh’s press conference. Stocks were already responding to an external shock that simultaneously worsened inflation and growth risks. The Fed’s cautious, noncommittal response did not reassure investors that it had an easy path through that shock. Markets were left to price both the economic damage from oil and the possibility of tighter policy.

The conflict dimension adds another layer of uncertainty. Energy prices can react not only to actual supply losses but also to the probability of disruption across shipping routes, export facilities and regional infrastructure. A diplomatic shift can reverse prices quickly; another escalation can push them higher. Monetary policymakers cannot predict those events with precision, which raises the value of scenario analysis and the cost of false confidence.

How Higher Long-Term Yields Pressure Stock Valuations

Stock valuation is often described through discounted cash flow. A company is worth the present value of the cash it may generate for owners. When the discount rate rises, the present value of distant cash flows falls. Long-duration growth stocks are especially sensitive because a larger share of their expected value lies many years ahead.

The Treasury yield is not the only component of the discount rate. Investors also require an equity risk premium, and individual companies carry business-specific risk. Yet the risk-free rate provides the foundation. A thirty-year Treasury yield above 5% changes the comparison between owning a bond and accepting uncertain corporate earnings. It also raises the financing cost for leveraged transactions, real estate and capital-intensive expansion.

This does not mean every yield increase produces an equal stock decline. If yields rise because growth expectations improve, stronger future earnings can offset the valuation effect. July 29 looked less benign because the yield move was tied partly to inflation, oil and policy uncertainty. Those forces can raise discount rates without producing an equivalent increase in real corporate growth.

Market concentration magnifies the effect. The largest technology companies carry substantial weights in broad indexes. When investors reduce the multiples assigned to AI-linked leaders, the S&P 500 and Nasdaq can fall even if many smaller companies move less. The Dow’s price-weighted construction creates a different pattern, but it too can suffer large point losses when high-priced components decline.

The correction in the Nasdaq 100 also matters psychologically. Once an index falls at least 10% from a recent high, the move is commonly labeled a correction. The label does not predict what happens next, but it changes positioning. Trend-following strategies may reduce exposure, options hedging can amplify moves, and investors who bought near the peak may sell into rebounds. A weak technical backdrop can make policy uncertainty more powerful.

Microsoft’s Quarter: Strong Demand, Strong Spending and Better Visibility

Microsoft reported revenue of $90.01 billion for its fiscal fourth quarter ended June 30, an increase of 18% from a year earlier. Operating income rose 18% to $40.60 billion. GAAP net income increased 31% to $35.77 billion, and diluted GAAP earnings per share rose 32% to $4.81. The quarter exceeded the consensus figures cited by major financial news organizations.

Azure and other cloud services revenue increased 43%. Microsoft Cloud revenue reached $59.3 billion, up 27%, while the Intelligent Cloud segment generated $39.3 billion, up 32%. Commercial remaining performance obligations rose 84% to $678 billion. RPO is not recognized revenue, but it provides a view of contracted business expected to be recorded over time.

The strength was not limited to a single frontier-model customer. On the earnings call, management said all sequential RPO growth came from customers outside frontier-model companies and that nearly 90% of full-year cloud revenue came from customers outside that group. That detail matters because it addresses concern that Microsoft’s AI growth depends too heavily on a small number of model developers.

Microsoft also disclosed that full-year Azure revenue exceeded $100 billion and that Microsoft 365 Copilot had more than 30 million paid seats. Those measures offer evidence of monetization beyond infrastructure. The company is selling compute, developer tools and productivity software into an enterprise distribution system built over decades.

The company’s cash generation remained substantial, though capital intensity was visible. Quarterly operating cash flow was $55.4 billion, up 30%, while free cash flow was $19.6 billion after capital expenditures. Microsoft’s official release also included discrete items: a $3.2 billion gain related to Anthropic, lower voluntary retirement expenses, severance costs and an Xbox impairment. Together, those items increased earnings per share by $0.27 relative to prior guidance.

The accounting details are important because headline earnings were not purely a measure of recurring cloud operations. The Anthropic gain and other discrete items affected net income. A careful analysis therefore separates operating momentum—Azure growth, cloud revenue, bookings and backlog—from one-time or non-operating influences.

After the close, Microsoft shares rose roughly 8% to 9% in late trading, according to Reuters and the Associated Press. After-hours moves can change before the next regular session and should not be treated as closing returns. Still, the response indicated that investors viewed the cloud results and guidance as strong enough to outweigh spending concerns, at least initially.

Microsoft guided to a fiscal first-quarter revenue midpoint around $90.4 billion and projected Azure growth of about 45% in constant currency. Guidance is a company forecast, not an achieved result. It nevertheless gave investors a more concrete bridge between current capital spending and near-term sales growth than many AI-linked companies have provided.

Meta’s Quarter Was Not a Revenue Collapse

Meta’s results require more nuance than the claim that the company reported “really bad” revenue and earnings. Revenue rose 28% to $60.80 billion, driven by a 14% increase in ad impressions and a 12% increase in the average price per ad. Daily active people across Meta’s family of apps increased 3% to 3.60 billion. Those are strong operating figures for a business of Meta’s size.

The weakness appeared below revenue. Costs and expenses increased 55% to $42.03 billion. Operating income fell 8% to $18.78 billion, and the operating margin narrowed to 31% from 43% a year earlier. Net income declined 14% to $15.85 billion, while diluted earnings per share fell 13% to $6.18. The quarter included a $2.4 billion legal charge and $1.18 billion of severance expense.

Capital expenditure reached $31.08 billion. Operating cash flow was $31.86 billion, but free cash flow was only $784 million, down from $8.55 billion a year earlier. Meta defines free cash flow as net cash provided by operating activities less purchases of property and equipment and principal payments on finance leases. The low figure therefore reflected enormous investment as well as movements in operating cash flow.

It is incorrect to treat the $7.77 billion year-over-year decline in free cash flow as a discrete amount Meta “spent trying to catch up with AI.” Free cash flow is a residual measure, not a line-item invoice. The comparison was affected by capital expenditure, legal and severance costs, taxes, working capital and other cash-flow components. Meta did spend heavily on infrastructure, but the precise AI-spending conclusion cannot be derived by subtracting the two free-cash-flow figures.

Meta raised the lower end of its 2026 capital-expenditure outlook to $130 billion, keeping the top end at $145 billion. It projected third-quarter revenue between $61 billion and $64 billion and full-year expenses of $165 billion to $169 billion. Those forecasts indicate management expects both continued revenue expansion and continued cost pressure.

The market’s initial after-hours response was negative. Meta shares fell about 4% immediately and at points were down closer to 10%, according to contemporaneous reports. The volatility reflected an argument over quality rather than the existence of growth. Investors had to decide whether a strong advertising engine justified the sharp increase in investment and the lower near-term margin.

The direct comparison with Microsoft can be misleading because their businesses differ. Microsoft sells cloud infrastructure and software subscriptions; Meta earns most of its revenue from advertising. AI can improve Meta’s recommendation systems, engagement, ad targeting and creative tools without appearing as a separate “AI revenue” line. The company’s 28% revenue growth and advertising metrics suggest AI-related improvements may already support the core business, even if the infrastructure bill is substantial.

The skeptical case remains serious. Meta’s free cash flow left little quarterly cushion after capital expenditures. Its legal exposure, severance costs and rising debt add complexity. Management must show that the infrastructure program improves revenue, efficiency or strategic control enough to earn an acceptable return. A strong ad market can temporarily mask the cost of a poorly disciplined investment cycle.

Microsoft and Meta: What the Comparison Really Shows

Sources: Microsoft and Meta quarterly earnings releases. Reporting periods and business models differ, so the comparison is directional rather than like-for-like.

The contrast is not “AI winner versus AI loser.” Both companies reported substantial demand and both are spending at historically high levels. Microsoft’s quarter offered more visible linkage between investment and revenue because Azure growth, commercial bookings and RPO quantified the demand. Meta’s linkage is more indirect because AI improves advertising and engagement while the infrastructure cost appears immediately.

Microsoft also benefits from selling the picks and shovels of the AI buildout. Customers pay for compute and software whether their own AI projects succeed or fail, subject to contract terms and usage. Meta is largely an end user of infrastructure, investing to improve products, protect strategic independence and develop new services. That can create enormous value, but the payback is less contractually visible.

Meta’s revenue growth was actually faster in percentage terms. The concern was the conversion of that revenue into operating profit and free cash flow. Microsoft’s revenue growth was slower but accompanied by stable operating-income growth and much higher cash generation after capital spending. Investors rewarded the clearer economics.

One quarter does not settle the strategic contest. Microsoft faces capacity constraints, dependence on enterprise budgets, competition from other cloud providers and the possibility that AI infrastructure becomes commoditized. Meta may convert its infrastructure into stronger recommendations, lower ad-creation costs, new products or proprietary models that improve margins later. The market reaction reflects the current evidence, not a final judgment.

Fact-Checking the Most Important Claims

Bitcoin Near $64,000: Resilience, Not Confirmation

Bitcoin traded near $64,000 around the Fed decision and the equity selloff. That relative stability was notable because risk assets were under pressure and the Nasdaq was falling. It suggested that the immediate selling impulse in cryptocurrency was less intense than in some technology shares. It did not establish that Bitcoin had decoupled from macro conditions.

Bitcoin reacts to several overlapping forces: global liquidity, real yields, the dollar, leverage, ETF flows, regulatory developments, miner economics and crypto-specific sentiment. A single session can reflect positioning rather than a durable change in fundamentals. The asset can hold one level while equities decline, then move sharply when leverage is liquidated or new information arrives.

The claim that a sideways pattern is “setting up for a major run” is technical interpretation, not verified fact. Chart patterns do not contain a guaranteed future outcome. A support level can attract buyers, but it can also fail. The responsible observation is that Bitcoin had not made a new local low during the immediate market reaction as of the research cutoff.

Higher long-term yields are not automatically positive for Bitcoin. A rising real return on government bonds can reduce demand for assets that produce no cash flow. Inflation anxiety can support the narrative that scarce digital assets are a hedge, but Bitcoin has also traded as a high-beta risk asset during periods of market stress. Which channel dominates changes over time.

Oil adds a further complication. If energy inflation keeps the Fed restrictive, liquidity-sensitive assets may face pressure. If investors lose confidence in fiscal or monetary stability, alternative stores of value may benefit. Those scenarios point in opposite directions, which is why a confident price forecast from one day’s chart is not justified.

The best conclusion is modest: Bitcoin’s ability to remain around $64,000 while major stock indexes fell showed short-term resilience. The next move will depend on broader liquidity, incoming inflation data, the dollar, Treasury yields and crypto-market positioning. Resilience is evidence; inevitability is not.

Why Microsoft Could Not Simply “Save” the Market

A strong Microsoft report can improve sentiment, but one company cannot neutralize all the forces that drove July 29. The market was contending with higher long yields, an energy shock, geopolitical risk and a correction in semiconductor shares. Microsoft’s after-hours advance supported index futures, yet the durability of that response depended on whether investors viewed the results as company-specific or representative of the broader AI economy.

Microsoft’s model is unusually well positioned for the current cycle. It owns a global cloud platform, enterprise software relationships, developer tools and distribution. Its Azure growth can therefore reflect spending by many industries. A chipmaker, consumer-device company or advertising platform may not receive the same benefit from the same level of AI enthusiasm.

The market also had other major reports still ahead, including Apple, Amazon, Coinbase and Strategy. Each tests a different part of the narrative: consumer-device demand, cloud and retail economics, cryptocurrency trading activity, and corporate Bitcoin exposure. Good results across several companies would provide broader confirmation than Microsoft alone.

Even excellent earnings do not erase the discount-rate effect. A company can raise its revenue outlook and still face valuation pressure if Treasury yields rise enough. The final stock price reflects both the expected cash flows and the rate used to value them. July’s market was repricing both sides of that equation.

The Strongest Bullish Interpretation

The constructive reading begins with the fact that the Fed did not hike. A majority judged that the existing rate range was appropriate despite the oil shock and elevated inflation. The labor market remained stable, June core CPI was flat for the month, and productivity investment was strong. If energy prices retreat and June PCE confirms disinflation, the committee may be able to remain on hold without causing a recession.

The yield-curve steepening can also be read as evidence that markets reduced the probability of an immediate policy mistake. Lower two-year yields suggest less expectation of a near-term hike. If long yields later settle as inflation fears ease, equities could recover without the Fed needing to cut. That would amount to a soft landing supported by productivity rather than monetary stimulus.

Microsoft’s results strengthen the case that AI spending is producing real revenue. Azure growth of 43%, a $678 billion commercial backlog and more than 30 million paid Microsoft 365 Copilot seats are not merely speculative promises. They show enterprises paying for infrastructure and software. If that demand persists, the investment cycle can support economic growth and corporate earnings.

Meta’s quarter also contained bullish evidence despite the cash-flow concern. Revenue rose 28%, ad impressions and pricing both improved, and user activity grew. If management can moderate spending or demonstrate that AI infrastructure lifts advertising efficiency, the current margin compression may prove temporary. The company’s core business remains highly profitable.

Bitcoin’s stability and the absence of broader credit-system stress are additional positives. The equity decline was sharp, but there was no evidence in the available data of a banking panic, frozen funding market or disorderly Treasury auction. A repricing can be painful without becoming a financial crisis.

The Strongest Skeptical Interpretation

The skeptical case focuses on credibility and timing. Inflation has remained above target for years, oil has surged again, and three policymakers already wanted higher rates. If the Fed waits too long, long-term inflation expectations and term premium may continue to rise. The central bank could then be forced to tighten into a weaker economy.

Warsh’s communication strategy may increase that risk. Reducing forward guidance can improve market discipline, but the Fed has not yet established a sufficiently clear replacement. If investors cannot infer the reaction function, every data release and geopolitical shock can produce larger moves. Higher volatility itself raises borrowing costs and can weaken investment.

The AI capital cycle is another concern. Microsoft’s results were strong, but the industry is spending hundreds of billions of dollars before the ultimate demand is known. Capacity constraints can support pricing today, then become oversupply if growth slows. The market may be extrapolating current cloud demand too far into the future.

Meta’s free-cash-flow compression demonstrates how quickly investment can consume operating cash. Even with 28% revenue growth, costs rose faster and the operating margin narrowed by twelve percentage points. If ad growth slows before infrastructure spending falls, the earnings pressure could become more severe.

Oil creates a scenario in which both consumers and valuations suffer. Higher energy costs reduce disposable income, while inflation keeps bond yields elevated. That combination can hurt revenue expectations and discount rates at the same time. It is one reason the July 29 selloff extended across several indexes rather than remaining confined to one company.

What the Market Is Really Debating

The core debate is not whether the Fed should have moved by exactly 25 basis points on one day. It is whether the economy’s supply capacity can expand fast enough to absorb AI investment, fiscal demand and energy shocks without persistent inflation. If productivity rises rapidly, current capital spending can increase real growth and make inflation easier to control. If productivity arrives slowly, the investment boom may add demand before it adds supply.

Markets are also debating the appropriate term premium for U.S. government debt. Long-dated Treasuries must absorb uncertainty about inflation, deficits, issuance and central-bank policy. A higher term premium can keep mortgage and corporate rates elevated even if the Fed eventually cuts the overnight rate. That weakens the familiar assumption that a pause automatically produces easier financial conditions.

Within technology, investors are separating revenue beneficiaries from spending beneficiaries and spending bearers. Chipmakers and cloud providers receive orders. Software companies must prove adoption. Platforms such as Meta spend to defend engagement and advertising. Utilities and construction firms supply the physical infrastructure. Each has a different margin profile and risk.

Cryptocurrency sits at the intersection of these debates. It can respond to liquidity and risk appetite like a technology asset, to inflation anxiety like a scarce asset, and to industry-specific flows like a separate market. Its behavior cannot be inferred from one macro narrative alone.

Implications for Households and Businesses

For households, the unchanged Fed rate does not mean borrowing costs will remain unchanged. Mortgage rates are closely tied to longer-term Treasury yields and mortgage-backed securities spreads. If the ten-year and thirty-year yields remain elevated, home financing can become more expensive even while the federal funds rate stays at 3.50% to 3.75%.

Credit-card and many variable-rate borrowing costs are more directly linked to short-term policy rates. A hold prevents an immediate increase in the benchmark, but it also delays relief. Consumers carrying revolving balances continue to face high interest expense. Higher gasoline prices add another burden, especially for lower-income households that spend a larger share of income on energy and transportation.

Businesses face a similar split. Companies with floating-rate debt care about the short end; companies issuing long-term bonds care about Treasury yields and credit spreads. A steepening curve can therefore tighten financing for capital projects even without a Fed hike. Highly leveraged businesses and commercial real estate are particularly sensitive.

For technology companies, the cost of capital affects the hurdle rate for data-center projects and acquisitions. Large cash-rich firms can self-finance, but suppliers and smaller developers may depend on external funding. If bond yields remain high, the AI investment boom could become more concentrated among the strongest balance sheets.

Energy-intensive industries face both financing and operating pressure. Airlines, trucking, chemicals and manufacturing may see higher fuel or feedstock costs. The ability to pass those costs through depends on demand and competitive conditions. Margin pressure can appear before official inflation data fully capture the shock.

What Would Justify a September Rate Hike

A September increase would become more likely if several conditions align. Inflation would need to show persistence beyond energy, labor demand would need to remain firm, and financial conditions would need to appear insufficiently restrictive. A further rise in core services inflation or a broadening of price increases would matter more than a temporary gasoline spike alone.

The Fed would also watch expectations. Survey measures, market-based inflation compensation and long-term Treasury yields can signal whether the public doubts the 2% objective. A rise in long yields is not automatically evidence of unanchored expectations because growth, issuance and term premium also matter. Policymakers would need to decompose the move.

Another factor is the balance sheet. Warsh has emphasized that the policy rate should not be considered in isolation. If the Fed believes its asset holdings or liquidity operations are providing more accommodation than intended, it could adjust those tools rather than relying entirely on the funds rate. The task-force review may influence that framework over time, though not necessarily by September.

A hike would carry risks. Monetary policy works with lags, and the labor market has already slowed. Raising rates into an oil shock can reduce demand without expanding energy supply. The committee would need confidence that the inflation risk outweighed the possibility of unnecessary damage to employment and investment.

What Would Support Another Hold

A second hold would be easier to justify if June PCE and subsequent July inflation reports confirm that underlying price pressure is easing. A retreat in oil prices or progress toward de-escalation would reduce the risk that headline inflation feeds into expectations. Softer employment or consumer spending would add reason for patience.

The Fed could also hold if financial conditions tighten through markets. Higher long-term yields, wider credit spreads and lower equity prices can reduce demand without a formal policy move. Central banks monitor those channels because the economy responds to the complete financial environment, not just the overnight rate.

Holding does not imply satisfaction with inflation. It can reflect a judgment that existing restraint is still working and that the value of additional information exceeds the benefit of immediate action. That is essentially the majority’s July position.

The communication challenge would remain. Another hold accompanied by vague guidance could keep term premium elevated. The Fed may need to explain the conditions for action more clearly even while refusing to pre-commit to a specific outcome.

The Data Calendar That Matters Next

The first major test arrives after this article’s cutoff. The Bureau of Economic Analysis scheduled the advance estimate of second-quarter gross domestic product and the June Personal Income and Outlays report for 8:30 a.m. EDT on July 30. The PCE price indexes within that release are central to the Fed’s inflation assessment.

The Bureau of Labor Statistics is scheduled to release the July employment report on August 7, the July Consumer Price Index on August 12 and the Producer Price Index on August 13. Those reports will show whether June’s softer inflation reading continued and whether payroll growth weakened further.

Energy prices will update continuously before those scheduled reports. If Brent remains near or above $90, economists will revise inflation and consumption assumptions. If the conflict de-escalates and oil reverses, the July shock may have a smaller effect on medium-term policy.

The next scheduled FOMC meeting is September 15–16. The Fed will release its statement at 2 p.m. EDT on September 16, followed by a press conference at 2:30 p.m. Market-implied probabilities will move before then, but they are snapshots of prices rather than forecasts with guaranteed accuracy.

Historical Perspective: Why Point Losses Need Context

Financial headlines often emphasize Dow point changes because the numbers are vivid. As the index level rises, however, the same percentage move produces a larger point change. A 2% decline when the Dow is at 50,000 equals 1,000 points; a 2% decline when it is at 10,000 equals 200 points. Point comparisons across decades therefore mislead unless they are converted into percentages.

July 29’s 2.2% Dow decline was severe but not comparable to the double-digit percentage losses associated with major crashes. The significance lies in the combination of drivers and the level of bond yields, not in the four-digit headline alone. It showed that a market near record levels remained vulnerable to simultaneous shocks in oil, rates and technology.

The day also differs from a classic recession panic. Economic activity was still described as solid, unemployment was 4.2%, and major companies were reporting strong revenue. Investors were repricing the price of risk and the credibility of future earnings, not responding to evidence that commerce had stopped.

That distinction matters for both journalism and portfolio interpretation. A correction can continue and become economically important, but the evidence should determine the label. “Selloff” accurately describes July 29. “Crash” would imply a scale or market dysfunction not established by the closing data alone.

The Federal Funds Rate Is Only One Part of Monetary Policy

The focus on whether the Fed raised or held the federal funds rate can obscure the rest of the monetary-policy system. The target range governs overnight lending between depository institutions, but the Fed also sets the interest rate paid on reserve balances, operates standing repo and reverse-repo facilities, and manages a large securities portfolio. Those tools influence liquidity, money-market rates and the transmission of policy through the financial system.

For the July decision, the Fed directed the New York Fed to maintain the federal funds rate within the 3.50% to 3.75% target range. The interest rate paid on reserve balances was set at 3.65%. The standing overnight repurchase facility offered funds at 3.75%, while the overnight reverse-repurchase rate remained 3.50%. The primary credit rate was 3.75%. These settings create a corridor that helps keep overnight rates near the committee’s intended level.

Reserve balances matter because banks use them to settle payments and meet liquidity needs. When reserves are abundant, the rate paid by the Fed helps establish a floor under money-market rates. When liquidity becomes scarce, repo facilities can prevent temporary funding stress from pushing rates far above the target. The mechanism is technical, but it determines whether the announced policy stance actually reaches financial markets.

The balance sheet affects the quantity and composition of reserves and the supply of duration held by private investors. When the Fed buys long-term Treasury or mortgage securities, it removes duration from the market and can lower term premium. When it allows securities to mature without reinvestment, private investors must absorb more duration. The resulting effect on long yields depends on issuance, demand, expectations and the pace of runoff.

Warsh has argued that the policy rate should not be viewed in isolation from the balance sheet. That is a significant departure from a simplified public debate in which every decision is reduced to “hike, hold or cut.” Two policy mixes can share the same federal funds rate while producing different financial conditions if the Fed’s asset holdings, reserve supply or liquidity operations differ.

The July 29 curve reaction illustrates the point. Short-term yields moved in response to expectations about the next rate decision. Long-term yields incorporated the expected future policy path, inflation risk, Treasury supply and term premium. The Fed’s overnight target remained unchanged, but the cost of long-duration financing rose. In practical terms, part of the tightening occurred outside the formal rate vote.

This broader perspective also explains why the Balance Sheet Policy Task Force could matter. The review may examine the long-run size and composition of Fed assets, the appropriate quantity of reserves, and the separation between monetary policy and market functioning. Any recommendation would need to be communicated carefully because balance-sheet changes can affect Treasury markets even when the funds rate is unchanged.

For investors, the lesson is to avoid treating the target rate as a complete measure of accommodation. Real yields, credit spreads, equity prices, the dollar, mortgage rates and lending standards all shape financial conditions. A central bank can pause while markets tighten, or raise its rate while other conditions partly offset the move. July 29 was an example of the former.

Why Economic Data Are Revised—and Why That Is Not Evidence of Manipulation

Economic reports are estimates assembled under time pressure. The first payroll, GDP or inflation release uses the information available at the deadline. Additional survey responses, tax records, company filings and seasonal data arrive later. Statistical agencies revise earlier estimates to incorporate that material. A revision can be large and consequential without being suspicious.

The monthly payroll report offers a clear example. The establishment survey samples employers and is later benchmarked against more comprehensive unemployment-insurance tax records. Initial responses are incomplete, especially around turning points and disruptions. Seasonal adjustment must separate recurring patterns—such as holiday hiring and school calendars—from genuine changes in employment. The result is useful, but not exact.

The household survey, which supplies the unemployment rate, measures people rather than payroll positions. A person with two jobs can appear twice in establishment payroll counts but once in the household survey. Self-employment is captured differently. Population controls can alter historical comparisons. Apparent conflicts between employment indicators often reflect definitions rather than dishonesty.

Inflation measurement has its own challenges. The CPI tracks a basket of consumer goods and services and updates weights to reflect spending patterns. Housing costs are represented partly through owners’ equivalent rent, an estimate of what homeowners would pay to rent their homes. Product substitutions, quality changes and new technology require statistical judgment. Analysts can reasonably debate those methods.

The PCE price index uses a broader consumption concept and different weights. It can account more quickly for substitution among categories and includes expenditures made on behalf of households. Because CPI and PCE answer slightly different questions, their inflation rates can differ. The Fed’s preference for PCE does not make CPI invalid; it reflects the policy objective and the properties officials value.

Private data introduce another layer. ADP processes payroll records from participating businesses and models broader employment. Online job postings, card transactions and mobility data can provide timely signals, but they cover different populations and can be affected by changes in customer mix or behavior. A private series may lead, lag or diverge from official data without either one being fabricated.

The proper response to uncertainty is transparency. Agencies should publish methodology, sampling error, revision history and known limitations. Researchers should compare multiple indicators and avoid false precision. Policymakers should describe how sensitive their decisions are to revisions. The Fed’s Data Task Force can contribute by identifying real-time signals and improving the way uncertainty is communicated.

The market relevance is substantial. If investors treat every first release as exact, prices can overreact. If they dismiss all official statistics as manipulated, they lose a disciplined basis for analysis. A better approach is to interpret each report as one estimate within a broader evidence set. Warsh’s statement that one month does not settle a five-year inflation problem is consistent with that approach.

The AI Supply Chain Is Larger Than Microsoft and Meta

The technology selloff cannot be understood solely through the earnings of two platform companies. The AI investment cycle spans semiconductor designers, memory manufacturers, foundries, networking suppliers, server assemblers, data-center developers, utilities, cooling systems, construction contractors, cloud providers and software companies. Demand can be strong at one layer while margins deteriorate at another.

Chip designers benefit when cloud and internet companies order accelerators, but they depend on manufacturing capacity and advanced packaging. Memory suppliers face a cyclical business in which scarcity can turn into oversupply. Networking vendors must keep pace with clusters that move extraordinary volumes of data. Utilities and grid operators need generation and transmission capacity. The final economics depend on the entire chain.

The July weakness in semiconductors showed that reported profit growth is not enough when expectations are extreme. A company can increase earnings and still disappoint if investors anticipated an even larger result. The distinction between absolute performance and performance relative to expectations is central in earnings season. High valuations create a narrow margin for execution.

Capital spending by hyperscalers appears as revenue for suppliers, but it is a cash outflow for the buyers. That transfer means the same investment can look bullish in one company’s income statement and risky in another’s cash-flow statement. Microsoft occupies both positions: it buys enormous quantities of infrastructure while selling cloud capacity. Meta is primarily a buyer whose return appears through advertising and product performance.

Power availability is becoming a binding constraint. Data centers require reliable electricity around the clock, and interconnection queues can delay projects. Companies may sign long-term power agreements, build near generation sources or invest in efficiency. Energy prices therefore connect the AI story to the Fed story. An oil shock does not directly determine data-center electricity costs, but broad energy scarcity and infrastructure constraints can raise the cost of expansion.

Construction and depreciation create another timing issue. Cash capital expenditure occurs when equipment and facilities are purchased. Accounting expense is recognized over the estimated useful life of those assets. A company can therefore report strong operating income while cash flow is pressured by investment. Later, depreciation expense rises even if capital spending slows. Investors must track both cash and accounting measures.

Useful-life assumptions matter. If specialized chips become obsolete faster than expected, the economic return can be lower than the depreciation schedule implies. If software and utilization extend asset life, returns can be higher. Companies have incentives to present investment as durable infrastructure, but the rate of technological change creates uncertainty.

Capacity utilization is equally important. A data center that operates near full utilization can generate attractive revenue. One built ahead of demand can become an expensive idle asset. Microsoft’s disclosure that Azure demand exceeded available capacity supports the case that current investment is needed. The question is whether that relationship persists as industry capacity expands.

Enterprise adoption determines the next stage. Pilot projects and experimentation create usage, but durable revenue requires applications that improve productivity, reduce cost or generate sales. Paid Copilot seats offer one measurable indicator. Cloud consumption and backlog offer others. The quality of adoption is stronger when customers renew, expand and integrate AI into core workflows rather than running temporary tests.

Meta’s economics are different because the return can be distributed across billions of users and advertisers. Better recommendations can increase time spent, improve ad inventory and raise conversion. Generative tools can reduce the cost of creating advertisements. These benefits may be real without a product labeled “AI” producing separate revenue. The challenge is proving that incremental profit exceeds infrastructure cost.

The market’s transition from enthusiasm to scrutiny is healthy. It does not mean the AI cycle is over. It means the burden of proof has shifted from technical capability to economic return. Companies with contracted demand, high utilization and strong cash flow will be treated differently from those relying on distant strategic promises.

Corporate Finance Under a Steeper Yield Curve

A steepening yield curve changes decisions far beyond public equity trading. Companies choose debt maturities based partly on the relationship between short and long rates. When long-term yields rise relative to short-term yields, issuing long-dated fixed-rate bonds becomes more expensive. Management may prefer shorter maturities, floating-rate debt or delayed financing, each of which creates a different risk.

Shorter debt can reduce the current coupon but increase refinancing exposure. A company that borrows for two years must return to the market sooner, possibly during a recession or credit event. Long-term debt locks in funding but may appear expensive if rates later fall. Treasury departments balance cost, duration, covenants and liquidity rather than simply choosing the lowest visible rate.

Credit spreads can amplify the Treasury move. A corporate bond yield equals a government benchmark plus compensation for default, liquidity and other risks. During an equity selloff, spreads often widen as investors become less willing to hold lower-quality debt. A company can therefore face a double increase: a higher Treasury yield and a larger spread.

Investment-grade technology companies with large cash balances remain better protected than leveraged issuers. Microsoft can finance much of its investment internally and access debt markets on favorable terms. Smaller data-center developers, suppliers and software companies may be more dependent on external capital. A higher long-term rate can concentrate market power by making scale and balance-sheet strength more valuable.

Private equity is particularly sensitive because acquisition models rely on debt and an assumed exit value. Higher financing costs reduce the price sponsors can pay while preserving target returns. They also lower valuation multiples if future buyers face the same expensive debt. A sustained rise in long yields can slow deal activity even without a recession.

Commercial real estate faces a direct refinancing challenge. Property values are often estimated by capitalizing net operating income at a market rate. Higher Treasury yields and risk premiums can raise capitalization rates, reducing estimated values. Owners refinancing maturing debt may need more equity or may be unable to borrow the previous amount.

Infrastructure projects and utilities also depend on long-duration financing. AI-related power generation and transmission require large upfront investments with returns spread over decades. A higher thirty-year benchmark increases the required revenue from those projects. That cost can eventually be reflected in electricity prices, cloud pricing or lower returns.

Household finance follows the same logic. Mortgage rates can remain elevated despite a Fed pause, reducing affordability and transaction volume. Auto and business loans respond through different benchmarks and credit conditions. The broad economic effect of July’s yield move will depend on whether it persists beyond a volatile session.

Market Causation: Why One Headline Is Rarely Enough

Financial markets move through the interaction of information, expectations and positioning. A news event can occur immediately before a price move without being the sole cause. The July 29 headline—stocks fall after the Fed holds—accurately describes sequence, but it does not prove that the hold caused every point of the decline.

Stocks were already falling before the policy announcement. Oil had surged, semiconductor weakness was spreading, and investors were preparing for major earnings reports. The Fed decision then changed the distribution of expected policy outcomes. Warsh’s press conference affected yields and confidence. The closing loss was the cumulative result.

Different assets emphasized different information. Energy markets responded primarily to geopolitical and inventory news. Two-year Treasuries reflected near-term policy expectations. Thirty-year bonds reflected longer-run inflation and term premium. Technology shares reflected discount rates and earnings risk. Bitcoin reflected crypto liquidity and broader risk sentiment. A single-cause story cannot explain all those movements.

Positioning also matters. If investors were heavily exposed to AI stocks, a modest disappointment could trigger disproportionate selling. Options dealers may hedge as prices move through strike levels. Systematic funds can reduce risk when volatility rises. These mechanisms do not replace fundamentals, but they influence the speed and size of the response.

The distinction between catalyst and cause is useful. The Fed press conference was a catalyst that intensified uncertainty. The underlying causes included inflation persistence, energy risk, valuation and market concentration. Strong journalism separates the immediate trigger from the conditions that made the market vulnerable.

This discipline also prevents overstatement about earnings. Microsoft’s report did not “cause” every after-hours gain, and Meta’s spending did not alone determine every decline. Prices incorporate guidance, expectations, positioning and the next day’s macro outlook. Precise language—shares rose after results, rather than because of one number—better reflects the evidence.

Scenario Analysis Through the September Meeting

Scenario One: Oil retreats and disinflation resumes

In the most constructive scenario, diplomatic progress reduces energy prices, June PCE confirms softer underlying inflation and July data continue the trend. Payroll growth remains positive but moderate. Long-term yields fall as inflation risk and term premium ease. The Fed holds in September and communicates that existing restraint is sufficient.

Under this scenario, technology shares could benefit from both lower discount rates and strong earnings. Microsoft’s results would look like evidence that AI demand remains durable, while Meta’s spending concerns would become more company-specific. Bitcoin could benefit from improved liquidity and risk appetite, though its response would still depend on crypto flows.

Scenario Two: Energy inflation persists but growth holds

If Brent remains elevated and economic activity stays solid, the Fed may face pressure to hike. The committee would need to decide whether the oil shock is temporary or broadening. Three July dissenters would have a stronger case, and additional members might join them if core measures reaccelerate.

Markets could experience continued curve volatility. Short yields would rise as a September increase became more likely, while long yields might remain high. Equity performance would depend heavily on whether earnings growth can offset the higher discount rate. Capital-intensive technology companies would face the greatest scrutiny.

Scenario Three: Oil stays high and employment weakens

This is the stagflation scenario. Headline inflation remains elevated while hiring and consumption deteriorate. The Fed’s choices become more painful because tightening would address expectations at the cost of additional labor weakness, while holding could appear tolerant of inflation.

Long-term yields might not fall as much as in a conventional slowdown if inflation risk remains high. Credit spreads could widen, producing tighter financing even without a policy move. Equity weakness would likely broaden from high-multiple technology into cyclical and consumer sectors.

Scenario Four: AI investment slows abruptly

A sudden reduction in hyperscaler spending would ease some demand and power pressure but damage semiconductor, construction and cloud-related earnings. The Fed might see less inflationary demand, yet the market could interpret the slowdown as evidence that prior investment was excessive.

Microsoft’s backlog and capacity commentary make an immediate collapse less likely, but industry expectations can change quickly. A few major buyers account for a large share of advanced-computing demand. Their capital budgets therefore have macroeconomic significance.

Scenario Five: Productivity accelerates faster than expected

The most favorable structural scenario is that AI investment quickly raises output per worker. Companies produce more without equivalent increases in labor and other inputs. Wage gains become compatible with lower unit costs, profit growth broadens, and inflation moderates despite solid demand.

This outcome would validate Warsh’s emphasis on productivity and reduce the need for severe monetary restraint. It is plausible but difficult to identify in real time. Productivity data are volatile, and the benefits of general-purpose technologies often arrive unevenly across industries.

Scenario Six: Communication uncertainty becomes the dominant risk

Even acceptable economic data could produce volatility if investors remain unsure how the Fed will react. Conflicting official speeches, political pressure or abrupt changes in market-implied probabilities could keep term premium high. The policy rate might remain unchanged while financial conditions tighten unpredictably.

The remedy would not necessarily be a return to calendar promises. The Fed could publish clearer scenario-based guidance, explain its balance-sheet framework and identify which evidence would change its risk assessment. Predictability of process matters more than certainty of outcome.

How to Read the Next Earnings Reports

The next wave of large-company results should be evaluated through more than revenue beats and after-hours percentage moves. Apple’s report will provide information about device demand, services growth and its ability to monetize AI features. Amazon will test both consumer spending and cloud demand. Coinbase will show trading activity and transaction economics in digital assets. Strategy will offer another view of corporate Bitcoin exposure and financing.

For each company, the first question is whether reported revenue is recurring, cyclical or influenced by one-time items. The second is whether margins improved or deteriorated. The third is whether operating cash flow converts accounting profit into cash. The fourth is how much capital expenditure is required to sustain growth.

Guidance deserves equal attention. A company can beat the completed quarter while warning that demand, costs or currency will worsen. Conversely, a modest historical result can be followed by an improving outlook. Consensus expectations should be dated because they change as analysts update models.

Balance-sheet strength will become more important if long yields stay high. Cash-rich companies can continue investing through a downturn. Heavily indebted companies may need to cut spending, issue equity or refinance at unfavorable rates. The AI cycle could therefore widen the gap between dominant platforms and smaller competitors.

Investors should also separate industry demand from market share. Strong cloud growth at one provider can reflect overall expansion, share gains or both. Weakness at one chipmaker can reflect product execution rather than a collapse in AI demand. Comparable periods, constant-currency figures and segment definitions matter.

Finally, after-hours moves are provisional. Liquidity is thinner, spreads can be wider and conference-call details can reverse the initial reaction. The regular-session close provides a more stable reference, though even that cannot prove a long-term conclusion from one quarter.

How to Interpret Market-Implied Fed Probabilities

After the July decision, futures pricing suggested roughly a 57% probability of a September rate increase at one point, down from higher levels earlier in the day. That number was useful as a snapshot of market prices, but it was not a poll of economists and not a declaration that a hike was more likely than not in an objective sense. It reflected the prices of short-term interest-rate contracts under a set of assumptions.

Tools such as CME FedWatch translate federal-funds futures into implied probabilities for possible target ranges. The calculation begins with the contract price, which reflects the market’s expected average effective federal funds rate over a calendar month. When a meeting occurs partway through that month, the expected average combines days before and after the decision. Analysts then infer the probabilities that would be consistent with the price.

The result is sensitive to the assumed set of outcomes. If the model considers only a hold and a 25-basis-point hike, it divides probability between those possibilities. If traders begin to price a larger move, an intermeeting action or unusual behavior in the effective rate, the simplified probabilities can become less informative. Technical factors and risk premiums can also affect futures prices.

Market-implied odds can change dramatically with one data release. A softer inflation report may reduce the expected rate; an oil spike or strong wage number may increase it. This flexibility is a feature because prices incorporate new information quickly. It is also a reason not to present one probability as a stable forecast.

The July curve contained richer information than the headline odds alone. The decline in two-year yields suggested less fear of an immediate hike, while the rise in long yields indicated greater concern about inflation or term premium over a longer horizon. A single September probability could not capture that tension. Investors were simultaneously pricing less near-term tightening and more expensive long-term money.

Policymakers do not vote according to futures markets. Warsh explicitly rejected the idea that the Fed simply follows market pricing. Markets can inform the committee about financial conditions and expectations, but the FOMC evaluates economic data, risks and its mandate. A large gap between market pricing and the eventual decision can still create volatility, which gives both sides an incentive to understand the other’s assumptions.

The practical use of implied probabilities is comparative. They help show how expectations changed before and after an event. If the September hike probability fell after the press conference, that indicates traders interpreted the event as less supportive of an immediate move than they had expected. It does not mean the long-run message was dovish, particularly when thirty-year yields rose.

Journalists and investors should therefore attach a date and time to probability figures and identify the source. They should avoid language such as “the market knows” or “the Fed will.” The prices summarize collective positioning under uncertainty. They are evidence of expectations, not evidence of the future.

Sector-by-Sector Effects of the July 29 Repricing

Technology and semiconductors

Technology faced the most direct combination of higher discount rates and questions about AI spending. Semiconductor companies are exposed to the capital budgets of a small group of enormous customers. Cloud and software companies must show that adoption produces recurring revenue. A higher long yield reduces the valuation of distant growth and raises the standard for an earnings beat.

Within the sector, balance sheets and business models matter. Microsoft’s cloud backlog offered visibility. Suppliers dependent on one product cycle or a few buyers may face more volatility. Consumer technology depends on device demand and replacement cycles, while advertising platforms depend on corporate marketing budgets. “Tech” is not one trade even when the index moves make it appear so.

Financials

A steeper curve can help banks if they earn more on longer-term assets than they pay on deposits and short-term funding. The benefit is not automatic. Higher long yields can reduce the value of bond portfolios, weaken loan demand and increase credit losses. Banks with large unrealized securities losses or concentrated commercial-real-estate exposure may be hurt rather than helped.

Trading operations can benefit from volatility, but wealth and investment-banking activity may slow if markets remain unstable. The net effect depends on funding structure, asset duration, credit quality and fee businesses. A simple claim that higher yields are good for banks ignores those differences.

Energy

Oil producers can benefit from higher crude prices through stronger revenue and cash flow, subject to hedging, production costs and taxes. Refiners may face complex margin effects, while airlines, transportation and chemical companies are generally exposed to higher fuel or feedstock costs. The energy sector can therefore outperform the broad market even as the oil shock worsens the macro outlook.

Geopolitical gains can reverse quickly. A ceasefire or restored supply can lower prices before producers change investment plans. Companies may return cash rather than expand drilling if they doubt the durability of the price. The sector’s response should be evaluated through realized prices and production, not spot crude alone.

Consumer discretionary and staples

Higher gasoline and borrowing costs reduce discretionary income. Retailers, restaurants, travel companies and automakers may see weaker demand, particularly among lower-income households. Consumer staples can be relatively defensive, but they also face transportation and packaging costs and may struggle to raise prices without losing volume.

The distributional effect matters. Wealthier households with financial assets may absorb energy increases more easily, while households living paycheck to paycheck must cut other spending. Aggregate consumption data can therefore remain stable while individual categories deteriorate.

Industrials and transportation

Industrial companies encounter both demand and input-cost effects. Defense and energy-infrastructure suppliers may receive stronger orders, while manufacturers and logistics providers pay more for fuel and financing. Companies with pricing power and long backlogs are better positioned than those operating on thin margins or fixed-price contracts.

Airlines are particularly sensitive because fuel is a major variable cost. Hedging can delay the impact, but sustained crude increases eventually reach jet-fuel prices. Shipping routes and insurance can also be affected by conflict, adding operational risk beyond the commodity price.

Real estate and utilities

Real estate is highly sensitive to long yields because property values and mortgage costs depend on duration. Residential affordability can weaken, while commercial owners face refinancing pressure. Data-center demand may support selected property types, but those projects require power and infrastructure and can be delayed by financing costs.

Utilities can benefit from growing electricity demand but must finance large capital programs. Regulators determine how quickly costs can be passed to customers. Higher bond yields can pressure utility valuations because the sector is often compared with fixed-income investments and carries substantial debt.

Small-cap companies

Smaller companies generally have less access to bond markets, thinner margins and greater dependence on banks or floating-rate loans. They can be more sensitive to a prolonged period of restrictive policy. The Russell 2000’s 1.6% decline on July 29 reflected broad risk aversion, even though the most visible headlines centered on large technology companies.

Small caps may benefit if the curve steepens because growth expectations improve, but July’s move was mixed with inflation and oil risk. Companies with domestic revenue may avoid currency exposure yet face direct pressure from U.S. wages, energy and financing. Balance-sheet screening becomes more important when refinancing conditions tighten.

Cryptocurrency-related equities

Exchanges, miners and companies holding Bitcoin can move differently from the token itself. Exchanges depend on trading volume and fee rates. Miners face electricity costs, equipment efficiency and block-reward economics. Corporate holders add leverage and financing structure to the underlying price exposure.

Bitcoin holding near $64,000 does not guarantee that all crypto-related shares will be stable. Equity investors also price management, dilution, debt, operating expenses and regulatory risk. The upcoming results from Coinbase and Strategy were therefore relevant tests of whether crypto-market resilience translated into corporate earnings and balance-sheet strength.

Risks and Uncertainties

Inflation persistence

The largest macro risk is that energy pressure spreads into broader prices while core inflation stops improving. Repeated shocks can change expectations and wage behavior. The Fed would then face pressure to tighten despite slowing employment.

Geopolitical escalation

The oil market remains sensitive to military developments and shipping routes. A disruption to production or transit could push prices higher quickly. Diplomatic progress could reverse the move just as rapidly, making forecasts unusually fragile.

Long-term Treasury supply and term premium

Long yields may remain high even if the Fed holds because investors are evaluating inflation, fiscal borrowing and duration risk. Persistent term-premium pressure would affect mortgages, corporate finance and equity valuation.

AI capital-expenditure discipline

Microsoft and Meta are spending at enormous scale. Demand is currently strong in several areas, but the industry must convert infrastructure into durable cash flow. Overbuilding, pricing pressure or slower adoption would weaken returns.

Market concentration and leverage

Broad indexes remain sensitive to a small group of large technology companies. Options positioning, systematic strategies and leveraged trades can amplify moves when volatility rises. A correction can therefore exceed the immediate change in fundamental value.

Data revisions

Employment, GDP and inflation data are revised. A policy decision based on incomplete real-time information can look different later. The Fed’s data review is relevant for this reason, but revisions are a normal feature of economic measurement rather than proof of manipulation.

Bitcoin liquidity

Bitcoin can move sharply outside regular U.S. equity hours. A stable price at one timestamp does not guarantee stability through the next data release or geopolitical development. Leverage and thin liquidity in parts of the crypto ecosystem can accelerate declines.

Frequently Asked Questions

Why did the Dow fall more than 1,000 points if the Fed did not raise rates?

The market was pricing future conditions, not only the current federal funds rate. Oil surged, technology shares weakened, three Fed officials voted for a hike, and long-term Treasury yields rose. The Fed’s hold removed an immediate increase but did not reduce uncertainty about inflation or September policy.

What did the Federal Reserve decide on July 29, 2026?

The FOMC kept the federal funds target range at 3.50% to 3.75%. Nine officials supported the decision, while Beth Hammack, Neel Kashkari and Lorie Logan preferred a 25-basis-point increase.

Was the July 29 decision a dovish hold?

Not in the conventional sense. The rate was unchanged, but inflation language remained firm and three officials dissented in favor of tighter policy. The result is better described as a cautious or hawkish hold.

Did Kevin Warsh end forward guidance?

Warsh has reduced the Fed’s reliance on prescriptive forward guidance and wants markets to respond more directly to data. The Fed has not stopped communicating; it still issues statements, holds press conferences and publishes projections. The change is one of emphasis and style.

Is the Fed investigating whether inflation data are rigged?

No such investigation was announced. The Fed created a Data Task Force to consider the quality and timeliness of economic information. Reviewing methodology and improving measurement is not an allegation that the Bureau of Labor Statistics, ADP or another provider manipulated figures.

Why did long-term Treasury yields rise while the two-year yield fell?

Traders reduced expectations for an immediate hike, which pressured the two-year yield lower. At the same time, investors demanded more compensation to hold long-dated bonds because of inflation, oil, policy uncertainty and term-premium risk. That produced a steeper yield curve.

Did Microsoft beat earnings expectations?

Microsoft’s fiscal fourth-quarter revenue and earnings exceeded the consensus figures cited by major financial news outlets. Revenue reached $90.01 billion, Azure grew 43%, and commercial remaining performance obligations rose to $678 billion. Some earnings were influenced by discrete items, including an Anthropic-related gain.

Were Meta’s earnings actually bad?

Meta’s revenue and advertising metrics were strong, but profit and cash-flow quality weakened. Revenue rose 28%, while costs rose 55%, operating income fell 8%, the operating margin narrowed, and free cash flow dropped to $784 million after heavy capital spending and other cash effects.

Did Meta spend nearly $8 billion to catch up in AI?

That conclusion cannot be calculated from the year-over-year free-cash-flow decline. Free cash flow is operating cash flow minus capital expenditure and related payments. Its change reflects multiple items, including infrastructure investment, legal charges, severance, taxes and working capital.

Why did Bitcoin hold near $64,000 during the stock selloff?

Bitcoin experienced less immediate selling pressure than major U.S. equity indexes during the measured window. That may reflect positioning, crypto-specific flows or different trading hours. It does not prove that Bitcoin has permanently decoupled from stocks or that a rally is certain.

When is the next Federal Reserve meeting?

The next scheduled FOMC meeting is September 15–16, 2026. The policy statement is scheduled for 2 p.m. EDT on September 16, followed by the chair’s press conference at 2:30 p.m.

What should markets watch before September?

The most important variables are the June PCE report, second-quarter GDP, July employment, July CPI and PPI, oil prices, long-term Treasury yields, credit conditions and further large-company earnings. No single release will necessarily determine the decision.

Research Cutoff and the Limits of the Snapshot

This analysis uses information available through 5:15 a.m. EDT on July 30, 2026. That boundary is important because several market-moving releases and regular-session price moves were still ahead. The Bureau of Economic Analysis had not yet published the advance estimate of second-quarter GDP or the June Personal Income and Outlays report. U.S. cash equity trading had not opened after the Microsoft and Meta announcements, so their late-trading moves remained provisional rather than official next-day closing returns.

Rapidly changing prices should be read as timestamped observations. Bitcoin near $64,000, Treasury yields reached during the July 29 session, and after-hours stock moves can all change without invalidating the underlying accounting or policy analysis. The durable evidence consists of the Fed’s official vote and statement, the companies’ reported quarterly figures, the published inflation and employment data, and the scheduled calendar. Market prices show how investors interpreted that evidence at a particular moment.

The distinction protects against hindsight. Later GDP, inflation, earnings or geopolitical developments may strengthen or weaken the scenarios described here, but they were not known at the cutoff. A current article should be updated with those releases rather than silently inserting them into the earlier decision context.

Final Assessment

The Dow fell 1,153 points after the Fed decision because an unchanged policy rate did not resolve the market’s central problems. Inflation remained above target, oil had surged, technology leadership was weakening and long-term Treasury yields were rising. The 9–3 vote showed meaningful disagreement inside the committee, while Warsh’s reduced reliance on forward guidance widened the range of outcomes investors had to price.

The strongest evidence supporting a calmer interpretation is that the Fed did not hike, core CPI had softened in June, the labor market remained stable rather than collapsing, and Microsoft produced concrete proof of cloud and AI demand. The selloff was not accompanied by evidence of a systemic funding crisis. A reversal in oil and continued disinflation could reduce the pressure quickly.

The strongest concern is that the economy is receiving an energy shock while long-term inflation compensation and term premium are rising. If the Fed’s communication framework remains unclear, financial conditions may tighten through bonds even without a formal increase. Technology companies must then justify extraordinary capital spending against a higher cost of capital.

Microsoft and Meta reinforce the same lesson from different directions. AI demand is real, but spending quality matters. Microsoft showed a clearer line from infrastructure to cloud revenue, backlog and cash flow. Meta showed robust advertising growth but also how legal costs, restructuring and infrastructure can consume cash. Neither quarter proves the final return on the AI buildout.

Bitcoin’s stability near $64,000 belongs in the evidence, but not as a forecast. It showed that the immediate equity selloff did not automatically produce a new crypto low. The next move remains exposed to yields, liquidity, the dollar and market leverage.

The July 29 session was therefore less about one Fed decision than about the price of uncertainty. Investors were asked to value long-duration technology profits, government debt and digital assets while the inflation outlook was being rewritten by energy and a new central-bank communication regime. The next data releases will narrow that uncertainty, but they will not eliminate the structural questions around productivity, capital spending and the long-run cost of money.

This article is provided for general informational purposes and does not constitute financial, investment, tax, or legal advice.

Sources

• Federal Reserve: FOMC statement, July 29, 2026
• Federal Reserve: Chair Kevin Warsh’s opening statement
• Federal Reserve: July 29 press-conference materials
• Federal Reserve: policy implementation note
• Federal Reserve: announcement of five policy task forces
• Federal Reserve: Kevin Warsh biography
• Reuters: analysis of the Fed decision and yield-curve reaction
• Reuters: Wall Street closes sharply lower after Fed hold
• Associated Press: July 29 U.S. stock-index closes
• Reuters: oil rises about 7% as Middle East airstrikes resume
• Bureau of Labor Statistics: Consumer Price Index, June 2026
• Bureau of Labor Statistics: Employment Situation, June 2026
• Bureau of Economic Analysis: Personal Income and Outlays, May 2026
• Bureau of Economic Analysis: first-quarter 2026 GDP advance estimate
• Bureau of Economic Analysis: release schedule
• Bureau of Labor Statistics: August 2026 release schedule
• Federal Reserve: September 2026 calendar
• Microsoft: fiscal fourth-quarter 2026 earnings release
• Microsoft: fiscal fourth-quarter 2026 earnings call
• Reuters: Microsoft cloud growth and after-hours reaction
• Meta Platforms: second-quarter 2026 earnings release
• Associated Press: Meta earnings and after-hours reaction
• CoinMarketCap: Bitcoin market data

Affiliate disclosure: Businessfinance.news may earn compensation from qualifying actions completed through selected links on this website, at no additional cost to the reader. Affiliate relationships do not influence our editorial reporting, analysis, or conclusions.