Capital Markets
Deals, listings and the cost of money
IPOs, mergers, buybacks and credit issuance — what the terms of a deal say about what this market is willing to fund right now.
2026-10-31
%22%2F%3E%0A%20%20%3Ccircle%20cx%3D%22220%22%20cy%3D%22150%22%20r%3D%22180%22%20fill%3D%22%2322d3ee%22%20fill-opacity%3D%220.12%22%2F%3E%0A%20%20%3Ccircle%20cx%3D%221010%22%20cy%3D%22130%22%20r%3D%22220%22%20fill%3D%22%2338bdf8%22%20fill-opacity%3D%220.08%22%2F%3E%0A%20%20%3Ccircle%20cx%3D%22920%22%20cy%3D%22520%22%20r%3D%22170%22%20fill%3D%22%2322d3ee%22%20fill-opacity%3D%220.06%22%2F%3E%0A%20%20%3Cpath%20d%3D%22M140%20430h920%22%20stroke%3D%22url(%23line)%22%20stroke-width%3D%2212%22%20stroke-linecap%3D%22round%22%2F%3E%0A%20%20%3Cpath%20d%3D%22M220%20360c120-150%20260-220%20380-220s260%2070%20380%20220%22%20fill%3D%22none%22%20stroke%3D%22%2367e8f9%22%20stroke-opacity%3D%220.5%22%20stroke-width%3D%2216%22%20stroke-linecap%3D%22round%22%20stroke-linejoin%3D%22round%22%2F%3E%0A%20%20%3Cpath%20d%3D%22M300%20398h600%22%20stroke%3D%22%2394a3b8%22%20stroke-opacity%3D%220.24%22%20stroke-width%3D%228%22%20stroke-linecap%3D%22round%22%2F%3E%0A%20%20%3Ctext%20x%3D%22220%22%20y%3D%22510%22%20fill%3D%22%23e2e8f0%22%20fill-opacity%3D%220.7%22%20font-family%3D%22Arial%2C%20Helvetica%2C%20sans-serif%22%20font-size%3D%2238%22%20letter-spacing%3D%228%22%3EINSIGHTS%3C%2Ftext%3E%0A%3C%2Fsvg%3E)
2026-10-30
Anthropic's $9.7 Billion Research Bill Is Why the Shorts Are Lined Up for the $2 Trillion Print
Anthropic's first-half research spend of $9.7 billion — up 427% year over year — turns its planned Nov. 16 Nasdaq debut into a collateral event for the whole AI-credit chain rather than a fund-raising event for the model business. With $131.7 billion of cash against $265.6 billion of non-cancelable purchase commitments and $417 billion of future minimum payment obligations, the equity becomes the funding plan that lenders, neoclouds and short sellers have been positioning against. A strong print de-risks that chain and forces the crowded short base in the most-shorted AI names to cover; a soft one re-prices $63 billion of loan collateral built on the assumption that GPU residual values hold. Allocation, not valuation, is where the risk sits over the next three weeks.
Binance's $6.3B Breach Is a Private Credit Event — and the Listed Crypto Proxies Traded It Like a Market Crash
About $6.3 billion left 17 Binance hot wallets across nine chains between 20:04 and 23:56 UTC on Oct. 29, 2026, and roughly $3.1 billion was still sitting in the attacker's wallets the next morning — unsold, and never routed to an exchange, a mixer or a bridge. That is what makes this the first exchange failure large enough to read as a credit question rather than a theft: the loss sits on one private balance sheet while the listed complex sold as if the market had been contaminated. The proxies are not mispriced on level so much as on composition, because the beneficiaries here are venues that capture migrating balances and balances that never reach a market, not the names carrying the most balance-sheet beta. Coinbase, Circle, Galaxy and Strategy each hold a different slice of that risk.
ExxonMobil's Q3 War Premium Is a Refining Story — and That Makes the Buyback a Ballot Issue
ExxonMobil releases third-quarter results before the open on October 30, and its own October 16 filing already shows where the war premium went: $4.5 billion of refining-margin uplift against a $1.3 billion upstream hit from higher crude prices. That mix matters politically, because the profit pool filling up is the one that sets the price at the pump, while the cash funding the buyback is running at 124% of trailing free cash flow. Democrats' Big Oil Windfall Profits Tax Act now names stock repurchases, so ExxonMobil's capital-return pace has become an election-month risk factor rather than a valuation lever.
Microsoft's $25 Billion Bond Ends the Net-Cash Myth: At 65 Basis Points, Its Spread Now Governs AI Capex
On October 29, 2026, Microsoft priced $25 billion of senior notes — its first US investment-grade bond sale since 2017 — with the 10-year at 65 basis points over Treasuries and a $115 billion order book. The bond does not add leverage so much as replace it: cash and short-term investments have already fallen $37.5 billion in twelve months while property and equipment rose $121.4 billion, so the $29.0 billion of net cash that headlines the balance sheet becomes $69.4 billion of net debt once $98.5 billion of lease obligations are counted. What changes is the price of money: with Microsoft's AAA rating setting the cheapest long-dated funds in the AI buildout, every marginal data center now clears a published 5.55% hurdle, and the spread — not the annual budget — decides how much capacity gets built. Investors should watch new-issue coupons on AI paper, because credit has not yet been asked the question equity asked on October 29.
2026-10-29
Ares Just Declined AI Chip Financing. The Money Still Exists — It Now Costs SOFR+550
On October 29, 2026, Ares Management chief executive Michael Arougheti said his firm would not underwrite financing against AI chips, because it cannot model how fast GPU residual values decay; Ares will concentrate on the roughly $2 trillion of data-center shells and power inside a greater-than-$5 trillion AI capital spending need through 2030. That is an underwriting standard, not a funding shortage: roughly $500 billion of AI-related debt was issued in 2026, the marginal compute-backed loan printed at SOFR+550, and 18 data-center financings now yield double digits. The gap is being filled by bank syndicates arranged by JPMorgan Chase and Mitsubishi UFJ Financial Group, by vendor and borrower balance sheets — including residual-value support of up to 25% from Nvidia, $30.2 billion of funded debt at CoreWeave and $5.1 billion of trailing interest expense at Oracle — and by fee-based platforms from Apollo Global Management, Blackstone, BlackRock, Brookfield Asset Management, Goldman Sachs and KKR, whose $500 billion headline is a mobilization target rather than committed capital. For investors, the shift matters less for who is bullish on AI than for where the residual risk now sits: outside the private-credit funds that built the AI lending playbook, and increasingly back on the vendor's own balance sheet.
26 Biotech IPOs, but the Float Doesn't Arrive Until April 2027
Biotech is the one sector still clearing the shut 2026 IPO window: 26 U.S. debuts, more than double 2025 and the highest count since 2021, yet the 22 offerings priced in the first nine months raised only about $7.5 billion, roughly 5% of what 114 U.S. IPOs raised through mid-October. The wave is genuine funding rather than sellers flipping stock, which is exactly why it cannot pay public shareholders before 2027: in the current queue just 18.8% of shares trade freely at listing and 79.9% become eligible 180 days later. The first live test already reads badly, with Kailera Therapeutics trading 16% below its $16.00 offer after its own mid-October lock-up expiry and a tolerability problem in its lead asset, and the 2020 cohort's base rate is 16% positive with 44% down at least 90%. Own the banks, the originators and the acquirers around the wave, not the listings inside it.
Cohere's CAD $35.7 Billion Is the Richest Revenue Multiple in AI: 106x Its Own Disclosed Sales, Funded by Governments
Cohere's Oct. 29 initial Series E close values the enterprise-AI lab at CAD $35.7 billion — about US$25.5 billion — which is roughly 60 to 95 times the recurring revenue implied by its own 165% growth rate, and two to four times what Mistral costs per dollar of sales on its own year-end target. The price was not set by model quality but by sovereign capital: Canada's government (up to C$1.6 billion of equity) and Schwarz Group (EUR 500 million) are underwriting a lab that has already contracted US$745.6 million of GPU capacity from listed landlords HIVE Digital and Boost Run. That chain turns a private mark into debt-financed data centers, which is where the credit risk now sits — and it leaves public AI software priced from 11.6x sales at Microsoft to 73.3x at Palantir, with no floor beneath it.
Disruptive's $10 Billion Blind Pool Just Became the Largest Bidder in the AI Power Bottleneck — and the Same Assets Trade at Half the Price in Public Markets
A Dallas firm that has never raised a traditional fund closed a $10 billion blind pool on Oct. 27, planning up to $1 billion checks into roughly ten late-stage AI, infrastructure and defense companies. The structural change is not the size but the concentration: one manager writing ten-plus mega-checks behaves like an infrastructure owner, buying the same scarce power, grid and industrial capacity that public companies from GE Vernova to Hyosung Heavy Industries are now selling at record backlogs. The contradiction investors must price is that the private marks on this cohort have fallen while the public marks on the same bottleneck assets have not — Groq's valuation dropped to $3.5 billion in a round Disruptive itself led, roughly a year after the $6.9 billion mark. Meanwhile GE Vernova's gas-turbine backlog and slot reservations reached 122 GW and Vertiv grew organic orders more than 100%. The bulk-buying shift puts private capital directly into the physical assets that public markets are repricing, and makes the public supply chain the cleaner way to own the thesis.
AI Bonds Just Hit a Record $419 Billion — and They Are Quietly Setting the Price of 30-Year Capital for Everyone Else
Global investment-grade and junk bond sales hit a record $6.61 trillion through Oct 28, 2026, up 15% from a year earlier, and AI-related US corporate issuance ran $419 billion, up 90% from full-year 2025. The crowding-out is real but it is not where most investors are looking: US investment-grade spreads ended August at 78 basis points, unchanged from the start of the year, because record foreign inflows and annuity demand absorbed the AI paper. What has repriced is duration — 41% of hyperscaler issuance now sits beyond 15 years, and the 30-year Treasury yield has risen about 40 basis points in 2026 to 5.25%. Alphabet can still fund the buildout out of $265.1 billion of cash, Goldman Sachs and JPMorgan get paid on the volume, and the credits squeezed first are the leveraged AI borrowers and the landlords financing the same megawatts.
The 2026 ETF Closure Wave Looks Like Passive “Forced Selling”—But It’s a Different Liquidity Test Than Investors Expect
In 2026, US ETF closures are running at an unusually high pace: the ETF.com closure tracker shows 148 listed ETF closures year-to-date, with more still possible as the year ends. While closures do create forced selling of underlying holdings, the bigger risk for market liquidity is whether the closure timing collides with already-thick AI/tech concentration—then compounds execution pressure through redemption-driven asset sales and lower underlying liquidity.
Upwind Turned Down Datadog's $4 Billion Bid and Printed $3.8 Billion Instead: AI-Security Now Clears at 42x ARR
Upwind closed a $400 million Series C at a $3.8 billion valuation on Oct 28, nine months after a $1.5 billion Series B, after rejecting a reported Datadog approach worth nearly $4 billion because it carried no premium over the round's own price. The mark values Upwind's roughly $90 million of exit-2026 ARR at about 42x — close to twice the 22.6x forward sales multiple implied by Datadog's own $100.8 billion enterprise value — so private capital, not a strategic buyer, now sets the clearing price for fast-growing cloud security. The read-through for public investors is uncomfortable: takeout optionality in mature observability (Elastic at 6.1x sales, Dynatrace at 8.3x) is effectively dead at private-mark prices, and the buyers rich enough to pay 40-80x ARR are hyperscalers with cheap currency, not the high-multiple software vendors pivoting into the category. Datadog's Q3 guide of $1.135-1.145 billion, or 28-29% growth against 36% in Q2, is the near-term test.
2026-10-28
Anthropic IPO “shutout” turns AI exposure into a proxy-price trade—start with ADRs, end with the next frontier listing
Bloomberg reports that investors shut out of Anthropic’s IPO allocation are rotating into “proxy” exposure trades across listed AI-linked companies and even crypto/derivatives. The mechanic matters: when direct access is rationed, marginal pricing shifts to the most liquid listed proxies—even when their fundamentals are only indirectly tied to the model’s success. The clearest market test case is how SK hynix’s US ADR listing traded at a large premium to its Seoul share during the early dislocation window.
Boston's Office Tower Sold Itself Back for 68% Less — and the Cheapest Mark on the Building Is the Only One Backed by Cash
On October 20, 2026, Rockpoint Group won the foreclosure auction of its own 32-story, 1.1 million square foot Boston tower with the only bid on the table: $260 million against the $806 million it paid in 2019. That 68% discount is real, but it is not a market price — the sole bidder was the defaulted owner, and the auction's $10 million deposit, W-9 requirement and 35-day closing were built to deliver the asset rather than to discover a price. The credit math is sharper than the headline: $260 million recovers roughly 55 cents on each dollar of the original $470 million mortgage, sitting right at the fat tail of single-borrower office loss severity, while Boston's own assessment still carries the tower at $425 million. For investors the sequence matters more than the auction: extensions are expiring, private credit and insurer balance sheets are absorbing the paper at 40 to 60 cents, and the public office landlords holding the assets — BXP, Vornado Realty Trust, SL Green Realty — are the ones left carrying marks built on the old numbers.
Indian Retail Can Now Buy Nvidia and SpaceX — but a ₹10 Lakh Tax Threshold, Not the $250,000 Cap, Sets the Ticket Size
Groww, Zerodha, Angel One and Upstox have turned US equities into a domestic app feature, and Indian households answered fast: outward remittances for equity and debt investment jumped 111% year over year to more than $321 million in August 2026. Annualised, the whole channel is worth about $3.9 billion — roughly $4 million of brokerage, 1.3% of Nvidia's trailing revenue, and what one percent of India's 46 million active broking clients would produce — so it cannot price a mega-cap, but it changes who holds one. The binding constraint is not the $250,000 remittance ceiling but the 20% tax collected at source above ₹10 lakh and the $60,000 US estate-tax exemption, which together cap the average Indian retail position. The investable consequence is narrow: own the names where that flow meets improving operating leverage, and treat the December 8, 2026 SpaceX lock-up expiry as a dated supply event.
Jio's Record Debut Puts $22.6B of Listed India Exposure on Alphabet and Meta — and Prices It at 8.5x Sales
India's largest-ever IPO opened at ₹1,358 on Oct 28, 2026, 21.4% above the ₹1,119 issue price, valuing Jio Platforms at ₹12.53 lakh crore (about $131.5 billion) and carrying the Reliance group past the Tata Group on day one. Alphabet and Meta together hold about 17.2% of the company after the issue — roughly $22.6B of listed value at the debut close, against the ₹77,311 crore they committed in the 2020 round that created those stakes. At 8.5x FY26 sales the market pays nearly Alphabet's own sales multiple for a telco whose customers pay ₹215.6 a month, which only works if 533.3 million subscribers are judged as a distribution rail for advertising and AI rather than as connectivity. For US investors the listing converts an illiquid, cost-based stake in the world's cheapest user base into a daily-mark asset, and puts a public price on every downstream rupee of Indian ad and AI spend.
US Margin-Debt Growth Fell 22 Points in Four Months. The Leverage Migrated to Interactive Brokers, Charles Schwab and Robinhood — Right as the Fed Hiked
CNBC's October 27 read — margin-debt growth decelerating from 53.7% in May to 31.4% in September — looks like deleveraging. It is not: over those same four months US margin debt rose 4.6% to $1.48 trillion, so every point of the \"slowdown\" came from the comparison base, not the balance. The leverage itself migrated: Schwab's, IBKR's and Robinhood's margin books grew 93%, 36% and 59% against 31.4% for the industry, lifting their share of US margin debt from 16.7% to 21.3% and creating roughly 36% of all new margin borrowing. That concentration arrives just as the Fed raised rates on September 16 and decides again on October 28 — and it has already gutted the economics of the business, with Robinhood's margin book up 59% while its net interest revenue fell 9%. Read the deceleration as a composition warning, not a top signal: the marginal levered dollar now sits at online brokerages, in AI hardware, and in foreign hands buying 3x semiconductor funds.
The First Public Price Tag on a Major-League Franchise: MSG Knickerbockers and MSG Rangers Trade 39% Below Private-Market Marks
Madison Square Garden Sports completed the carve-out of the New York Knicks and New York Rangers at 11:59 p.m. ET on Oct. 26, 2026, and the first two trading days revalued nothing: the two new companies together are worth $9.44 billion against $9.36 billion for the combined parent a week earlier, and roughly $10.4 billion of enterprise value against $16.95 billion of private-market marks. The separation published the long-standing 39% discount rather than curing it, because a public price tag can only ever mark a minority, non-controlling, arena-leased interest. Two mechanics decide whether that gap closes: the local rights fees cut 28% for the Knicks and 18% for the Rangers now expire in 2028-29 and must be re-auctioned against two separate counterparties, and the league is selling the same schedule to Kalshi and Polymarket with no disclosed pass-through to franchises. For privately held teams the template reads as a haircut — the first public benchmark for a major-league franchise is what a non-controlling slice fetches, and that is roughly 61 cents on the dollar of private marks.
$2.1 Billion Bought Factories, Not Trucks: US EV Capital Moved Up a Layer, and Rivian's Cash Burn Explains Why
On October 27, 2026, Re:Build, the manufacturing platform that spun out Slate Auto, announced $2.1 billion in new funding co-led by General Catalyst and JPMorgan Chase's Strategic Investment Group — money for industrial capacity and new industrial startups, not for a vehicle program. That is the working template for US EV capital now: plants and fleets get capitalized, product ventures get funded against volume, and no OEM parent balance sheet sits in the middle. The weak point is the volume itself. Slate's roughly 180,000 reservations are refundable deposits rather than purchase contracts, while the same structure at Lucid Group is underwritten by Uber's commitment to at least 35,000 vehicles. Rivian's balance sheet is the clearest read on who is winning and who is paying: $697 million of venture marks and a $506 million non-cash gain against $3.49 billion of trailing free cash burn and a $1.0 billion loan at 6.03% secured on the joint venture.
SK hynix Turned Record Profits Into a $25.6B Payday — and the Same Shares Cost 44% More in New York Than in Seoul
Third-quarter operating profit of KRW 83.3 trillion at an 80% margin let SK hynix declare KRW 34 trillion — about $25.6 billion — of dividends and buybacks in a single board meeting, inside a roughly KRW 100 trillion FY2026 return target that is funded by cash generation rather than borrowing. The tension sits in the venue: ten ADRs equal one Seoul share, and at the Oct 27 Nasdaq close the American line priced the identical equity about 44% above the Oct 28 Seoul close, because Korean rules cap new ADR creation at roughly 2.5% of the share register and block the arbitrage that would close it. For investors the question is no longer whether AI memory converts into cash but which line they pay that cash in — and whether a 44% scarcity premium survives a payout cycle.
What to expect
Evidence-first notes with a visible point of view.
This section collects sharp takes on earnings, shareholder meetings, and market structure. Each new piece should make the thesis, the facts, and the implications obvious within the first few screens.
Expect direct analysis, not generic commentary.
Expect the data to be explicit and the argument to be easy to follow.