Warren Buffett, 96, stepped down Friday as chairman of Berkshire Hathaway, closing out a 60-year run that turned a failing New England textile mill into a $1 trillion conglomerate. Under his stewardship, Berkshire compounded shareholder returns at 19.7% annually, nearly double the S&P 500’s pace over the same six decades. Buffett now becomes chairman emeritus and stays on as a director; his son Howard Buffett takes over as chairman under a long-standing succession plan, while CEO Greg Abel, who assumed operating control nine months ago, continues running the business day to day.

The timing is the real story for investors. Berkshire shares are up just 1% in 2026 while the S&P 500 has rallied more than 11%, and the company is still sitting on a $365.5 billion cash hoard that shareholders have been waiting to see put to work. Abel has started answering that question: buybacks jumped to $4.5 billion last quarter, and Buffett revealed in July that he personally pushed for Berkshire’s $10 billion private purchase of Alphabet stock in June. That single bet now makes Google’s parent the conglomerate’s third-largest equity holding, trailing only Apple and American Express.

None of this changes Berkshire’s underlying playbook: buy durable, well-moated businesses at sensible prices and let retained earnings and insurance float compound for decades. What changes is who now owns full accountability for executing it. Abel inherits $44.5 billion in annual operating earnings, nearly 400,000 employees, and a market expecting him to match a track record widely regarded as the best in investing history.

So what for long-term investors: Berkshire’s lagging 2026 performance is a useful reminder that even history’s most disciplined capital allocators go through stretches where patience underperforms a hot, momentum-driven market. The title change in Omaha matters less than the substance behind it — watch whether Abel keeps shrinking the share count and finding genuinely undervalued businesses, because that discipline, not the chairman’s name on the letterhead, is what actually produced the 19.7% number in the first place.

Credit markets are starting to price in a risk that stock investors have mostly shrugged off: the debt-fueled AI buildout at the four largest hyperscalers is getting riskier, not safer. Apollo Global Management’s chief economist Torsten Slok flagged this week that credit default swaps on hyperscaler bonds — essentially insurance against a default — have widened by roughly 60 basis points relative to the banks that underwrite their debt since October 2025. That gap had sat near zero for years. For patient investors, this is worth watching closely, because bond markets have a long history of smelling trouble in balance sheets before equity markets do.

The numbers explain the unease. Alphabet is running a forward debt-to-equity ratio of 13% alongside negative free cash flow of $25.7 billion. Amazon’s debt-to-equity sits at 23% with free cash flow of negative $30 billion. Meta’s leverage is highest at 34% debt-to-equity, also with $25.7 billion in negative free cash flow. Microsoft is the outlier of the group, at just 7.3% debt-to-equity and a positive $33.4 billion in free cash flow — a reminder that not all AI capex stories carry the same balance-sheet risk. Slok’s point isn’t that these companies can’t service their debt today; it’s that CDS investors, some of the most sophisticated credit analysts anywhere, are attaching materially more risk to a debt-financed capex cycle with uncertain payback timelines on rapidly depreciating chips and data centers.

Bulls have a reasonable counter: margins at the hyperscalers are already inflecting positive, and the payoff from this capacity may not show up cleanly until 2027 or 2028. That’s a fair rebuttal, but it also means today’s negative free cash flow at three of the four largest AI spenders isn’t a temporary blip — it’s a multi-year bet that revenue conversion accelerates faster than depreciation and interest costs compound. So what for long-term investors: this isn’t a signal to abandon mega-cap tech, but it is a signal to differentiate. Companies funding AI investment out of existing cash flow, like Microsoft, carry a fundamentally different risk profile than those leaning on debt markets to chase the same opportunity. Watching credit spreads, not just earnings headlines, may be the more useful early-warning system for whether this capex cycle stays disciplined or starts cutting corners.

Copper just did something it hasn’t done since 1994: it strung together its longest weekly winning streak on record, with London Metal Exchange prices touching an all-time high near $14,700 a ton this month. The headline number matters less than what’s driving it — a structural supply gap that has almost nothing to do with speculation and everything to do with decades of underinvestment colliding with the AI buildout’s insatiable appetite for wire, cabling, and power delivery.

The math is stark. A single AI data center consumes roughly ten times the copper of a traditional one, and that demand is layering onto a mining industry that takes seven to ten years to bring a new project online — a timeline no amount of capital can compress. Meanwhile, the refining side of the business has quietly become a domestic scarcity story of its own: the U.S. operated 16 primary copper smelters in 1976. Today there are just two, with a third mothballed. That collapse in refining capacity means roughly a third of the copper mined in America gets shipped overseas for processing before being bought back as finished metal — a round trip that leaves the country’s AI and grid-modernization ambitions dependent on foreign refiners.

That concentration is precisely why this looks less like a commodity trade and more like a moat. Companies that already run functioning U.S. smelters — chiefly Freeport-McMoRan, up roughly 44% year-to-date — sit on the scarce end of a market where treatment and refining charges have collapsed toward zero, squeezing processors everywhere except the handful with pricing power. Existing mines are aging, ore grades are declining, and BlackRock’s thematic investing chief recently described the installed base bluntly as “tired, very, very old assets.” None of that reverses quickly, regardless of where prices sit next quarter.

For patient investors, the interesting question isn’t whether copper spikes or dips in the next few weeks — it already has, sliding nearly 6% off its September high as tariff clarity wavered. It’s whether the underlying supply-demand imbalance persists for years, which the mine-development timeline suggests it will. Diversified miners and royalty companies with copper exposure, alongside domestic refiners insulated by scarce smelting capacity, offer a way to own a multi-year structural theme rather than chase a headline price move. So what for long-term investors: when a critical input to the defining technology of the decade faces a supply chain that can’t be rebuilt on a politician’s timetable, the companies sitting on the choke points — not the traders chasing the daily print — are the ones worth watching.

The 10-year Treasury yield broke above 5% this week for only the second time since the 2008 financial crisis, and the move matters far more to long-term investors than another daily headline about bond math. The last time this threshold gave way was October 2023; before that, you have to go back to April 2007 — months before the Great Financial Crisis. The yield has now climbed 76 basis points this year alone, dragging the average 30-year mortgage rate to 6.76% and pushing the yield on the ICE BofA High Yield corporate bond index to 7.42%, up 89 basis points since January. Every dollar of future corporate earnings is now being discounted at a meaningfully higher rate than it was twelve months ago.

Strategists are split on what that means, and the disagreement itself is instructive. Barclays calls 5% a “historically important inflection point,” warning that beyond it, rising rates tend to become a persistent headwind for stocks rather than a side effect of a strong economy. ING’s Padhraic Garvey says a further run to 6% — last seen in 2000 — “would cause stresses” and “could potentially cause the risk asset space to fall over.” BlackRock’s Investment Institute takes the other side, arguing that higher yields only become a problem when they’re driven by fiscal panic rather than genuine growth and productivity gains; for now, it’s keeping its equity and AI overweights intact. CNBC’s Mike Santoli adds a sobering historical echo: the Fed hiked into a 10-year yield near a round-number threshold in 1999 too, amid a tech capex boom that later unwound painfully.

The practical distinction for portfolios is which companies actually feel this. Businesses funding heavy capital spending — including the AI hyperscalers plowing record sums into data centers — face a rising cost of that capital just as investors start demanding proof of returns on it. Meanwhile, companies with durable free cash flow, pricing power, and self-funded balance sheets are largely insulated from the discount-rate math punishing high-multiple growth stories. Dividend payers with strong coverage ratios also become relatively more attractive once “risk-free” cash yields north of 5% raises the bar every other asset has to clear.

So what for long-term investors: a market where capital finally has a price again rewards balance-sheet discipline and penalizes companies whose valuations depend on multiple expansion rather than earnings growth. This is the moment to stress-test whether your holdings’ competitive moats and cash generation can carry them if borrowing costs stay elevated — because “higher for longer” is no longer a hypothetical.

Three unglamorous dividend stocks just picked up high-conviction buy calls from Wall Street’s top-ranked analysts, and the mix — pipeline infrastructure, upstream oil, and a regulated utility — hints at where patient income investors might find durable value while the broader market obsesses over AI valuations.

Energy Transfer, the pipeline giant spanning 44 states, now pays a 6.3% annualized yield on its $1.36-per-unit distribution. JPMorgan’s Jeremy Tonet, ranked among the top analysts tracked by TipRanks, raised his price target after the company lifted 2026 EBITDA guidance to as much as $19.1 billion, up from a prior ceiling of $18.6 billion, while tightening capital spending to a $5.6-$5.9 billion range. That combination — rising cash flow funding a well-covered payout rather than one propped up by borrowing — is exactly what long-term income holders should be screening for.

Permian Resources offers a smaller 2.7% yield but a more interesting growth story. Management’s bolt-on acquisition strategy, dubbed the “ground game,” has already closed $1.05 billion in deals this year, and Goldman Sachs projects free cash flow per share compounding at a 20% annual rate through 2028. For dividend-growth investors willing to trade current yield for a rising payout down the road, that trajectory matters more than today’s headline number.

The most contrarian call is Sempra Energy, upgraded to Buy by Jefferies despite trading at a 14% discount to its electric-utility peers. The discount reflects real risk — a contested Texas transmission buildout and lingering California regulatory uncertainty — but the analyst argues the market has already priced in the worst case, calling the stock an early opportunity with limited further downside. Sempra’s 3.1% yield isn’t really the story here; the valuation gap is.

So what for long-term investors: none of these three require calling a market top or bottom. Energy Transfer offers cash-flow-backed income today, Permian Resources offers dividend growth funded by disciplined M&A, and Sempra offers a valuation gap that could close once regulatory clouds lift. Spreading an income allocation across yield, growth, and value within the same sleeve tends to hold up better than chasing whatever payout looks biggest this particular quarter.

Amazon’s cloud business just posted its fastest growth in nearly five years, and the smart money is paying attention. Goldman Sachs added 5.5 million shares of Micron and over 465,000 shares of Amazon in its latest 13F filing, while billionaire investors Stanley Druckenmiller, Peter Thiel, Seth Klarman, and David Tepper all built or expanded Amazon positions during the second quarter. Amazon is now the single most widely held stock among tracked billionaire portfolios — 62 of them, up from 59 the quarter before. That’s not a momentum trade; it’s a long-term conviction signal from investors who typically hold for years, not weeks.

The number driving that conviction: AWS revenue grew roughly 37% year-over-year in the second quarter, the fastest pace in 18 quarters, as AI workloads filled data center capacity faster than Wall Street modeled. Less discussed but arguably more important is what’s happening underneath the cloud number. Amazon’s custom silicon business — the Trainium and Graviton chip lines that reduce reliance on Nvidia and cut AI infrastructure costs — has already crossed a $25 billion annualized revenue run rate and is compounding at triple-digit percentage growth. That’s a vertically integrated moat few competitors can match: Amazon isn’t just renting out compute, it’s designing and selling the chips underneath it.

For patient investors, the real debate is about capital discipline, not growth. Amazon expects roughly $220 billion in capital expenditures this year, and free cash flow has already compressed under that spending. Management’s counterargument is that servers and networking hardware typically break even in under three years, and much of its AI capacity is pre-contracted for multi-year terms — a structural difference from speculative buildouts elsewhere in the sector. If that math holds, today’s margin pressure converts into a multi-year earnings and cash-flow acceleration once the current investment cycle matures. If AI demand cools before that payback window closes, the capex bill comes due with less to show for it.

Advertising and logistics automation add a second and third growth engine that rarely make headlines but continue to expand at a steady clip, giving Amazon more than one lever if cloud growth normalizes. So what for long-term investors: this isn’t a story about chasing an AI headline, it’s about whether a company with three compounding, cash-generative businesses can absorb a historically large capex cycle without permanently impairing returns on capital. The 13F data suggests some of the market’s most patient, valuation-conscious investors are betting it can.

Six years ago, Warren Buffett stood in front of Berkshire Hathaway shareholders and admitted he had been “simply too optimistic.” Berkshire had just taken an $11 billion write-down on Precision Castparts, the complex-metal-components maker it bought for $37.2 billion in 2016 — Buffett’s biggest acquisition ever, and by his own account, one where he paid “a very high multiple.” The pandemic gutted the aerospace industry, Precision Castparts’ largest customer base, and the deal looked like a rare Buffett misstep.

It’s taking the better part of a decade, but the thesis he never abandoned — that Precision Castparts was “the best in its business,” run by a CEO he trusted — is now playing out in full. A shortage of the turbine blades and complex components the company makes has emerged, driven by both a recovering aerospace sector and a new source of demand: natural gas turbines needed to power AI data centers. This week, GE Aerospace agreed to pay $11.75 billion for Consolidated Precision Products, one of the only real competitors to Precision Castparts. Barron’s calls that price “pricey” at 26 times projected 2027 EBITDA — but using that same multiple, it estimates Precision Castparts alone could be worth roughly $100 billion. That’s nearly triple what Berkshire paid for the entire company nine years ago, and well above the $60–75 billion estimate Barron’s floated just last month.

The kicker for long-term investors: none of this appears priced into Berkshire’s stock. Analyst Andrew Bary notes Berkshire isn’t “getting much credit” for the subsidiary’s ballooning value, partly because CEO Greg Abel — like Buffett before him — skips the investor-relations theater of analyst calls and glossy conference appearances that might otherwise draw attention to a division quietly worth multiples of its purchase price.

So what for long-term investors: this is a case study in the difference between price and value playing out in real time. Buffett didn’t sell Precision Castparts after the write-down, didn’t chase a quick exit to save face, and didn’t let a short-term miscalculation override a long-term read on quality and management. Nine years and one pandemic later, the “mistake” is on track to be one of Berkshire’s better-performing acquisitions by dollar value created — a reminder that patience and conviction in a well-run business can outlast even a very public admission of having overpaid.

Jane Street isn’t known for making noise. The New York trading giant, which oversaw more than $1 trillion in 13F securities as of the second quarter, rarely telegraphs conviction the way a hedge fund manager might. So when its latest filing revealed two brand-new positions — one in an early-stage nuclear reactor developer, the other in a European energy major paying a near-5% dividend — it’s worth understanding why a firm built on quantitative precision decided both belong in the same basket.

The smaller, riskier bet is X-Energy, a roughly $71 million stake in a company building small modular reactors, or SMRs. X-Energy doesn’t sell electricity; it sells the reactor technology, engineering, and licensing that let others build nuclear plants faster and cheaper than traditional utility-scale projects. The U.S. Department of Energy has already committed up to $2.1 billion toward its first commercial project with Dow in Texas, and Amazon is reportedly working with X-Energy toward more than 5 gigawatts of new nuclear capacity to power data centers. That’s a real signal: the AI buildout’s insatiable appetite for electricity is pulling nuclear back into serious institutional portfolios after decades on the sidelines. The catch is patience — commercial reactors are still years from operation, and licensing delays or cost overruns are the norm, not the exception, in nuclear construction.

The second position tells a different story. Jane Street put roughly $94 million into TotalEnergies, the French oil-and-gas major that currently yields close to 4.7%, a payout well covered by earnings. Unlike X-Energy, TotalEnergies is a cash-generating machine today, not a promise about tomorrow. Pairing a speculative nuclear moonshot with a durable dividend payer looks less like a bet on any single energy source and more like a bet that the world’s energy transition will be messier, slower, and more capital-intensive than the clean narratives suggest — rewarding investors who own both the disruptors and the incumbents funding the interim.

So what for long-term investors: nuclear power’s return to the institutional conversation, driven by AI data-center demand, is a multi-decade theme worth watching rather than chasing — but the smarter entry point may be the established energy majors already generating the cash flow and dividends to fund that transition while collecting a paycheck in the meantime.

Birkenstock just handed long-term investors a rare data point in a brutal season for retail: a footwear brand that raised its full-year guidance instead of walking it back. Fiscal third-quarter revenue climbed 15% in constant currency to about $829 million, beating estimates, and management now targets 15% full-year growth — the top end of its prior range — while lifting its adjusted EBITDA floor to at least 710 million euros. Shares jumped as much as 20% on the news, but the more durable story sits beneath the pop: this is a company whose customers are simply refusing to trade down.

That resilience shows up everywhere in the numbers. The Americas grew 14%, Europe 15%, and Asia-Pacific a striking 23% — all in constant currency, all without discounting. Management attributes the strength to affluent buyers who haven’t pulled back the way the broader consumer has, meaning Birkenstock is capturing pricing power precisely where most apparel and footwear names are losing it. Adjusted EPS rose 19% year-over-year, helped along by a $250 million accelerated buyback that trimmed the share count by roughly six million shares — a capital-allocation choice that rewards patient holders rather than chasing growth for its own sake.

It isn’t a flawless print. Adjusted EPS of 0.74 euros missed the 0.76 euro consensus, and currency swings plus U.S. tariffs weighed on margins, offset only partly by manufacturing scale. Hedge fund ownership also dipped from 32 funds to 25 between the first and second quarters, suggesting institutions grew cautious right before the resilience became obvious — a reminder that sentiment often lags fundamentals at inflection points. The open question for shareholders now is whether that Asia-Pacific growth rate, still the smallest of the three regions in absolute dollars, can compound into a genuine second leg of expansion alongside the mature European and American businesses.

So what for long-term investors: Birkenstock is demonstrating something increasingly rare in consumer discretionary — a brand moat strong enough to preserve full pricing during a spending slowdown, paired with disciplined buybacks that convert steady growth into faster per-share earnings gains. Investors watching for durable competitive advantages in a choppy retail environment now have a fresh data point that brand loyalty, not just price, still wins wallet share.

ASML Holding (ASML) and Taiwan Semiconductor Manufacturing (TSM) just handed long-term investors something rarer than a hot quarter: a multi-year roadmap. On September 8, the two companies confirmed that TSMC will begin using ASML’s next-generation High-NA extreme ultraviolet lithography systems in high-volume advanced-node manufacturing starting in 2030, with a follow-on transition to larger 12-inch photomasks targeted for 2033. That’s a six-year runway, not a headline trade — and it’s exactly the kind of visibility patient capital should pay attention to.

The two stocks carry different risk profiles inside the same roadmap. ASML gets paid when equipment ships and gets accepted, regardless of whether TSMC’s chips ultimately hit target yields. TSMC, meanwhile, only converts this technology bet into shareholder value if the new masks lower costs, lift throughput, and support pricing that justifies the capital outlay — a bigger if, playing out over a longer horizon. Neither company disclosed order values or earnings contributions, so this is best read as improved demand visibility, not a confirmed revenue bump.

Ownership data suggests institutional money is already leaning in quietly. Insider Monkey’s tracked hedge fund sample counted 140 ASML holders in Q2 2026, up from 133 in Q1 — roughly a 5% increase — while TSMC holders climbed to 249 from 234, up about 6%. Short interest stayed tame heading into the announcement: just 0.34% of ASML’s float and 0.59% of TSMC’s as of mid-August, signaling little skepticism baked into either stock. Both companies also carry founder-grade competitive moats: ASML is the sole global supplier of extreme ultraviolet lithography tools, and TSMC manufactures the overwhelming majority of the world’s most advanced chips.

So what for long-term investors: this is a reminder that semiconductor capital cycles run in decades, not quarters. A 2030-to-2033 rollout means today’s ASML and TSMC shareholders are underwriting years of execution risk before any payoff shows up in earnings. That’s uncomfortable for anyone chasing quarterly catalysts, but it’s precisely the kind of durable, moat-protected, multi-year compounding setup that rewards investors willing to hold through the boring middle innings.