# Special Situations Report — Weekly Call (readable notes)

- **Title:** A hawkish Jackson Hole and a Sunday-night Iran escalation bookend the week — Warsh recommits to the 2% PCE target and refuses forward guidance, taking September hike odds to 50-70% and December to 90%, while Iranian missiles hit a US air base in Jordan and WTI jumps 1.5% to $84.50 with Brent near $89.40; Bessent's doubled long-end buybacks and the floated $950B TGA drawdown are dismissed by the bond market and attacked by Druckenmiller in the WSJ as Treasury-run QE, with US debt through $40 trillion and debt service at 20% of tax revenue; Nvidia's $96.2B / +106% quarter, $279B of commitments and a 70%-through-FY28 growth guide add a record $453B of market cap in a day and then give it back on chip tariffs, hawkish Fed and circular-financing scrutiny; hyperscaler CDS at all-time wides against $3.1 trillion of off-balance-sheet arrangements; the software relief rally (CRM, CRWD, OKTA) against Intuit's guide-down; and three actionable special situations — the **Arbutus (ABUS)** modified Dutch-auction odd-lot tender at $5.00-5.75 expiring September 29, the **PRTH** activist campaign to force Tom Priore's $6.00 take-private toward $7.50-8.00, and **PayPal (PYPL)** after Stripe walked, traded 51.65 → 54.40 with a published valuation model and a plan to re-buy near $50
- **Show:** Weekly SSR research call (premium subscriber recording — no public video)
- **Guest/host:** Jay Singh (founder, Special Situations Report; ex-Goldman Sachs)
- **Date:** 2026-AUG-30 (Sunday call; the deck/transcript title it "2026-08-30 - SSN")
- **Source PDFs (this folder):** `transcript.pdf` (the verbatim premium transcript, its section headers carrying the recording's `MM:SS` cues) and `report.pdf` (the condensed report + actionable items + the economic/earnings calendar + the stock-symbol list) and `agenda.pdf` (the 49-page weekly deck he walks page by page). Also circulated on the call and saved here: `Citi Mkt Update Volatile Markets.pdf`, `HSBC MultiAsset Insights.pdf`, `Merrill on US Balance Sheet and Yen.pdf`, `Standard Chartered Global Outlook.pdf`, and his own published `PYPL Valuation and Risk Framework.xlsx` (uploaded live during the call to the SSR Q&A tab).
- **Note:** readable notes extracted from the premium PDFs; fillers (um/uh/you know/false starts/stutters) removed, wording otherwise verbatim from the recording. Section headers carry the recording's `MM:SS` cue — the source writes them as continuous minutes past the hour (65:26), normalised here and in the analysis page's Ref column to `1:05:26`. No public video, so the per-name table has no deep-links. Auto-transcript garbles corrected to the right entity and listed here: **"SSN" = SSR** (Special Situations Report — the transcript and report titles both garble it); **"Besen" / "Bessant" = Scott Bessent** (Treasury Secretary); **"Worsh" = Kevin Warsh** (Fed Chair); **"P Sky" / "P-Sky" = PSKY (Paramount Skydance)**; **"Arbitus" = Arbutus Biopharma (ABUS)**; **"Rmax" = RE/MAX Holdings (RMAX)**; **"the KV KV transaction" = the Kenvue (KVUE) / Kimberly-Clark (KMB) deal**; **"the ECS" = the EC**, the European Commission; **"IRN" = IREN** (formerly Iris Energy); **"Jevons's Paradox" = the Jevons paradox**; **"Sema Shah" = Seema Shah** (Principal Asset Management); **"Termino" = Project Tetromino**, Amazon's automated-delivery-station effort; **"StarMind" = Starmind**, the SpaceXAI satellite; **"Steamboat Partners" = Steamboat Capital Partners**; **"Q&R tab" = the Q&A tab**; **"Druck" = Stanley Druckenmiller**; **"the City piece" = the Citi piece**; **"empirical research in New York Morgan" = Morgan Stanley research** (the report attributes the 70%-of-EPS-growth estimate to Morgan Stanley). Internal errors and inconsistencies in the source, left as spoken and flagged here: the OpenAI/Broadcom inference chip is called **"Jalapeño" in the weekly summary and "Palomino" in the Tuesday recap** — the report writes it "Palomino/Jalapeño"; both refer to the same part; **"Nvidia's net revenue was 60.1 billion versus an estimate of 67 billion"** should read *US* revenue ($60.1B vs a $67.5B estimate), not net revenue — total revenue was $96.2B; **"consensus expects 70% uh 30% growth in 2027"** is a live self-correction (30% is the figure); **"MicroStrategy… 1.59 billion in new US dollar cash reserves"** and later **"increased its US Treasury reserve to 5 billion, established an additional 1.6 billion of cash"** are the same disclosure stated twice with different rounding (the deck: USD Reserve to $5.10B plus $1.59B of new USD cash); **PRTH's net debt is given as "950"** against the deck's $899.7M, and the earnings that preceded the bid are again called **"Q3"** for a company on a December year-end (the deck says Q2 2026); **"the offer is between 5 and 5.75… you would effectively make 42 cents because it closed at 5.09 after hours on Friday… but if you did it at close at 5.07, you make 43 cents"** is a live re-computation (the deck prices the trade off the August 25 close of $5.20, giving −1.8% at the $5.00 floor and +12.9% at the $5.75 cap); **"the odd-lot priority… shareholders owning 99 shares"** is stated as "99 shares" where the tender document's condition is 99 shares *or fewer*; **"Warsh's speech… we might, we must be confident"** is a self-correction inside the quote; and **"IGV is still up only 4% year to date"** and **"IGV has rallied 25%"** describe different windows (year-to-date versus since June 22). The report's stock-symbol list contains three tickers — **AMAT (Applied Materials), FSLR (First Solar) and GILD (Gilead)** — that appear nowhere in the recording, the report body or the deck; they are ignored here rather than invented into rows. The transcript footer's "Date of Transcript: 2024-05-23" and its "ssn_transcript_v6 - 2026-01-26" filename are stale template artifacts — the real call date is **2026-08-30** (Sunday).

---

2026-08-30 - SSR - Hawkish Jackson Hole, Iran Escalation, NVDA Earnings, PRTH, ABUS Odd Lot, PYPL Bid Fails

## Geopolitical Escalations & Market Impact - 00:00

Alright, so happy Sunday everyone. Hope you guys are doing well. As I mentioned earlier, not the best way to start the week with Iran escalations this Sunday. A rather unfortunate way to start the week.

Just today we had bombings for a US Air Force base in Jordan, and Iran launched those missiles at the US air base in Jordan, following strikes near the Strait of Hormuz a few hours ago. So over the weekend you have seen a sharp move in the price of oil.

If you look at oil futures and interest rate futures, what you're seeing is US markets have taken a leg lower. We are 30 basis points lower on the S&P 500, 40 basis points lower in the Nasdaq, 20 basis points lower in the small cap index. And while bond yields are flat right now, they'll probably be up by tomorrow morning. US oil is up a percent and a half to 84.50 for WTI. It was down earlier, and then Brent is almost at 90, 89.40, also up.

It's surprising to me that the US continues to do this bombing right ahead of midterm elections, like we are in September. I don't think anyone had anticipated these bombings happening in September and October of this year, but this is the unpredictable geopolitical environment that we're in. So we just have to accept it and navigate the markets as we see them here.

So we'll revisit this in a moment.

## Upcoming Earnings Calendar - 01:59

But in terms of the earnings calendar, again, we're 90% done with earnings season. We'll spend a lot of time analyzing earnings season on this call. But the most important earnings next week on a go-forward basis are Dell on the AI side, Palo Alto Networks, cybersecurity. CrowdStrike and other cybersecurity names, Okta, etcetera, did really well last week. GitLab this week potentially could beat just given there's so much more code that's coming out in the market. And I think people are just using GitLab more than ever. It's kind of like the Jevons paradox. The cheaper coding becomes, the more you use it, and GitLab is a way to record it.

So we'll see how that plays out. I don't have a position right now. We have Broadcom next week. Based on Nvidia's outlook, you would think that Broadcom would beat and raise its outlook as well. Nvidia's looking at 70% revenue growth. And again, some of that revenue growth is contingent on financing for Anthropic doing an IPO in October and OpenAI securing financing. And the reason why Nvidia credit spreads have continued to widen out, and the reason why Nvidia stock didn't rally — it rallied a lot, it rallied about 400 billion in market cap in the first day, over 7% — the reason why it fell the next day is number one, Warsh's hawkish speech, which we'll talk about in a moment. And then number two, a lot of Nvidia's growth is actually being financed by itself, right? Through these circular capital arrangements. So I think people are not as confident, even though the number is spectacular, that if the credit markets unravel whether that number can actually be hit. And that's an uncertainty at the moment.

It's not the base case, but it's still a big uncertainty.

Now, the economic calendar is a lot more busy than the earnings calendar. The most important economic events next week are the US manufacturing PMI, which is coming out on the first of September. It's supposed to be modestly positive at 53.3 versus 53.2. You have ISM manufacturing where prices paid is supposed to come down a little bit. Construction spending, JOLTS or job openings which are supposed to fall. ADP employment, factory orders, durable goods, S&P US services PMI, and the change in non-farm payrolls.

## Weekly Market Summary & AI Sector Performance - 04:35

Now on the following page, we do a weekly summary of the most important events during the week and then later, around page 30, we go through individual daily market-moving events, but let's quickly summarize this. So equity performance was split across factor lines this week following Fed Chair Kevin Warsh's hawkish Jackson Hole address. Large-cap value led the tape with the Dow advancing 50 bips and the S&P also around 50 bips for the week. More specifically, Dow was a little bit better, 53 bips, S&P 47 bips.

Conversely, because of the interest rate volatility with the two-year spiking, small caps lagged significantly, shedding about 140 bips as duration and rate-sensitive names sold off at the end of the week.

Remember that the Russell has a lot — half the companies are unprofitable. So if interest rates go up and they have debt, that affects them more than large-cap companies. So that is why it's quite relevant.

One of the themes of last week was the AI infrastructure and semiconductor ecosystem earnings. So Nvidia had a very strong quarter, as we talked about, delivered massive Q2 results, 96.2 billion in revenue, up 106% year-over-year, driven by 89 billion in data center revenue and a surge in backlog commitments to 279 billion, up 119 billion quarter-over-quarter. The CFO of Nvidia, Colette Kress, projected 70% revenue growth through fiscal year 2028, noting full-scale production of the Vera Rubin and Vera CPUs. Following the print, Nvidia added 453 billion in market cap in a single session, the largest one-day market cap gain in US history.

In terms of expanded alliances and ecosystem reach, AWS (Amazon) announced a major multi-year expansion to deploy 2 million additional Nvidia GPUs across its global infrastructure in 2027-2028. Hugging Face — Nvidia reportedly agreed to acquire Hugging Face for 12.9 billion, consolidating its moat across developer distribution, software tooling, and inference deployment. This will definitely help build the CUDA infrastructure. Then we have SpaceX, which outlined plans to deploy Nvidia-powered Vera Rubin NVL72 architectures in orbit via its first-generation Starmind AI satellite, which is much earlier than anticipated.

## Custom Chips, Software, and Frontier Models - 07:00

And in terms of custom chips and supply chain realignment, OpenAI did some internal testing showing that its Broadcom-built Jalapeño inference chip outperformed Nvidia's GB300 in throughput per watt and latency at 700 watts, with deployment targeted for later this year. We're optimistic for Broadcom results this week, but we'll see. Marvell upgraded its long-term outlook, projecting FY28 revenue to reach 18 billion, up 50% year-over-year, but the stock still didn't rally given this was priced in.

Apple is evaluating DRAM from CXMT and NAND from YMTC, two China-based memory companies, which pressured traditional memory players like SanDisk, down 6% during the week, Micron down 5% during the week, and that was in addition to US potential chip sanctions and tariffs, which is why memory underperformed AI overall.

In terms of software, we had post-Nvidia earnings a lot of quality software names, especially the cybersecurity names, rally very sharply. Okta was up 19%, Salesforce was up 12%, CrowdStrike was up 10% on record 333 million net new ARR, demonstrating that core security and enterprise workflows remain resilient. Conversely, Intuit fell 3.2% after guiding fiscal year '27 EPS significantly below consensus, as the risk of AI disruption in tax seems to be a bigger issue.

## Frontier Models, Data Centers, and Macro Trends - 08:28

In terms of frontier model and platform shifts, Anthropic is preparing for a public listing that could target a $2 trillion valuation and seek up to $100 billion in capital, more than SpaceX, pitching a $30 trillion total addressable market, which is bananas because US GDP is only $32 trillion. Meta is reportedly spending up to $10 billion annually on Anthropic models to power its internal and external agents.

In terms of data and observability, ClickHouse crossed 350 million in ARR. It's up 40% since May, driven by massive AI agent data loads, including OpenAI scaling to 30 petabytes a day, which is unbelievable.

In terms of data center compute and conversion, IREN detailed a pivot from Bitcoin mining to AI compute, reporting 1 billion in operating ARR and securing 2.8 billion in GPU debt financing. The stock still didn't really outperform.

In terms of macro, monetary policy, and fiscal dynamics, there is Fed tightening risk back on the table. We had an in-line US core PCE at 3.3% year-over-year, alongside firm consumer spending, which pushed short-dated yields higher, along with the Jackson Hole speech, with money markets fully pricing in a December rate hike. I think there's like a 70% chance of an October rate hike, which I think is bizarre, given I don't understand why Warsh would raise rates ahead of the midterms, one month ahead of it. Trump would be furious. But the markets are also pricing in a 90% probability of a December rate hike. Second quarter US GDP matched expectations at 1.5% annualized, while corporate profits surged 22.8% year-over-year to a record 4.8 trillion, or 14.9% of GDP.

## Bond Market Structure & Global Central Bank Actions - 10:06

In terms of the bond market structure and duration debates, Treasury Secretary Scott Bessent raised the minimum long-end bond buybacks from 2 billion to 4 billion per operation, with officials indicating that the $950 billion Treasury General Account, or TGA, could be tapped for additional firepower. Investor Stanley Druckenmiller publicly criticized the plan in a Wall Street Journal op-ed, arguing that suppressing long-term yields removes duration risk and mimics quantitative easing by the Treasury rather than the Fed.

## Global Central Bank Actions & Geopolitical Risk - 10:35

Global central bank hawkishness: the Bank of Korea raised rates by 25 bips to 3%, its second hike this year, while European Central Bank sources signaled potential September tightening amidst strengthening German export confidence.

In terms of energy, geopolitics, and policy risk — Iran sanctions and the Strait of Hormuz standoff. Treasury Secretary Bessent announced a zero-leakage enforcement campaign targeting Iranian digital assets, gold, aviation, and shipping. Geopolitical volatility flared as Iran's IRGC conditioned the reopening of the Strait of Hormuz on foreign warships remaining 400 kilometers away, offsetting lower risk premiums from reported US-Iran diplomatic channels.

Hedge funds executed their largest weekly purchase of global energy equities in nearly four years, shifting to their most overweight positioning relative to global equities since June of 2024, which means they probably anticipated the escalation this weekend. I wouldn't be surprised if there were some leaks from the executive branch about these bombings that are happening over the weekend.

## Trade Policy and Institutional Flows - 11:40

In terms of trade policy and tariff escalations, Canada announced retaliatory tariffs on US goods, raising steel duties by 50%, while the White House evaluates new semiconductor duties that could extend to downstream hardware like servers and laptops. The US also put 50% tariffs on Canadian steel and put tariffs on 20 billion of additional Canadian products, including hockey sticks.

Now, in terms of institutional flows, regulatory action, and crypto treasuries, there's a big equity de-grossing and gold inflows before the hawkish Friday with Jackson Hole. Hedge funds registered their largest weekly net sell of US equity since April of 2025, heavily selling stocks in tech, industrials, and utilities. Conversely, physical and gold ETFs saw 6.4 billion in weekly inflow, the third largest weekly inflow in history, driving gold total ETF AUM to 615 billion, so almost two-thirds of a trillion.

## SEC Scrutiny and Corporate Holdings - 12:45

SEC scrutiny on leveraged AI bets. The SEC issued subpoenas to major prime brokers including Goldman Sachs, JPMorgan, Citigroup, and Bank of America regarding margin calls and risk management surrounding the hedge fund Situational Awareness, which suffered a 67% portfolio drawdown in July following concentrated leveraged tech losses due to 4 times leverage, and then ended up selling the entire portfolio to Citadel, which then flipped 80% of the portfolio in the first week.

In terms of corporate treasury Bitcoin holdings, MicroStrategy expanded its capital framework, establishing 1.59 billion in new US dollar cash reserves while increasing its total treasury holdings to roughly 4% of circulating Bitcoin supply at near-zero net leverage after raising all this cash.

## Tech Commentary (Dan Niles) - 13:37

In terms of tech commentary, Dan Niles gave his weekly tech commentary today, and it's summarized below. So last week, SPX, Nasdaq, Russell — we were plus 50 bips, plus 0.8 percent, and negative 1.5% with oil down 4%. But a hawkish Warsh on Friday led to a bear flattening of the yield curve and a sell-off in the IWM.

Despite Nvidia's guide of 70% revenue growth versus consensus of only 47%, the semiconductor index ended down 2.3% while software was up 5.9% on solid earnings. In general, many AI investors have been bullish on semiconductors and bearish on software on the belief that AI will displace many point-solution software companies. This is why the SOX index is up 62% year to date, and IGV is still up only 4% year to date versus the S&P up 13%. Situational Awareness was the poster child for this type of positioning. But since the unwinding of the momentum trade, which started on June 22nd — he wrote about this concern on June 20th — IGV has rallied 25%, which is unbelievable, while SOX is actually down 22%.

For perspective, the Morgan Stanley Momentum Index — momentum long performance minus momentum short performance — from June 22nd to August 28th is actually down 36%, even after its bounce in August. And the more concentrated TMT index is down 54%. But a bullish twist on AI for the software sector introduced recently is that AI agents will access software tools 10 to 100 times more often than humans. On August 6th, Atlassian, which was in the bucket of software names widely considered at risk of being replaced by AI, rallied 35% the next day in reaction to solid earnings and outlook.

Then on August 13th, Workday rallied 18% on the news that private equity firm Silver Lake might be pursuing an acquisition, which probably put a temporary floor underneath software. Workday was also supposed to be in the AI crosshairs, and private equity has higher bars to clear given their use of leverage and holding period than the typical investor. Then on August 26th, Salesforce reported very solid results and guidance and a deal with Anthropic, in which they also first invested in May of 2023. The stock was up 23% in reaction the next day. This seemed to be a strong counterpoint to the SaaS-pocalypse worries. This strategic alliance with Anthropic allows users to execute actions natively inside Claude without needing to open traditional software screens. Salesforce also seemed to be changing how they charge customers, with fees more related to customer use and benefits to their business rather than per user. Then on August 27th, Workday reported results which were good enough, but arguably acquisition prospects drove more of the stock reaction to positive 6% the next day.

Historically, systems of record, security, and gaming software have been the only three areas he's liked. He now wonders whether the fundamental implications of Atlassian, Workday, and Salesforce are supportive of the technical reactions in the software stocks as a group, as agentic AI continues to ramp.

So how does he square this with his concerns that the rapidly escalating amount spent on AI by corporations has to come from somewhere? Annualized revenue run rates for Anthropic and OpenAI have ramped from 29 billion at the start of the year to 105 billion just seven months later, most of the growth coming from Anthropic. Software spending globally, excluding AI, was roughly 1 trillion in 2025, but IT services at 1.7 trillion is a bigger category which I believe still has risk. And finally, knowledge worker compensation is an even bigger category where disruption would be even less noticeable, at an estimated 35 to 50 trillion in 2025 for roughly 30% of the global workforce.

Looking forward, the deal on Friday for Venezuelan oil fields that hold the largest crude reserves in the world at 17 to 18% of global oil reserves should give a positive start to the week with declining oil prices. Now that has flipped with the missile launches in the Middle East, but we'll see how it plays out at the beginning of the week.

But a bit further out, don't fight the Fed, given he believes a hike is likely on 9/16 because the October 28th meeting is right before midterms. I tend to disagree. I don't necessarily think we're going to see an interest rate hike on September 16th unless oil goes a lot higher. Two, September has the poorest seasonality of all months. And three, there's even worse seasonality than normal during midterm election years. Four, recent bipartisan pushback against data center expansion, one of the few things both sides seem to agree on, tends to be a big risk on the AI infrastructure names. As Warren Buffett says, the market has to keep pitching you, but you don't need to swing.

## Special Situation Update (M&A) - 18:43

Now, in terms of special situation update, there's actually a couple of things that we want to buy. We found a new odd-lot tender which is quite interesting. We'll talk about that in a moment. But before we do that, let's just talk about some of the big M&A situations from the past week.

Shares of Boston Scientific were trading down 3% in the pre-market on Wednesday after the company disclosed a cybersecurity incident affecting certain of its information technology systems that has resulted in global disruption to the company's operations.

Warner Brothers traded firm on Tuesday after several media outlets reported that Paramount has been considering structural remedies including separate distribution agreements with cable operators. There's no set date as of now for when the state AGs and Paramount Skydance will sit down for settlement talks again. In other news, the EC commenced its review of the Kenvue / Kimberly-Clark transaction with the phase one deadline set for September 29th. Caesars will host its shareholder vote on September 22nd, and VCTR announced that it has entered into a definitive agreement to acquire First Eagle for about 7 billion, comprising approximately 4.4 billion in cash and 2 billion of new Victory Capital equity.

Shares of Warner Brothers rose to their highest level since February on Monday, after several reports over the weekend indicated that the California State AG and Paramount were set to hold settlement talks on Monday. California Attorney General Rob Bonta called off the scheduled meeting with Paramount executives afterwards, citing allegations that the company's leadership has leaked information and negotiated in bad faith amid ongoing discussions over a potential settlement.

"Not only did Paramount leak the alleged substance of settlement discussions, but they misrepresented these discussions, demonstrating a lack of good faith," Bonta said in a statement. "As soon as Paramount stops playing games and engages more sincerely, my office is happy to meet again." According to the Wall Street Journal, state AGs were preparing to ask Paramount Skydance to divest some cable channels as well as commit to keeping its movie studio separate from Warner Brothers.

In other news, Two Harbors announced that it has received final regulatory approval and the merger is scheduled to close on August 25th, which it did. Two Harbors stockholders of record at the close of business will be entitled to receive a stub period dividend in an amount equal to 20 cents per share of TWO stock. The stub period dividend will be paid with the merger consideration and will not reduce or otherwise affect the merger consideration. RE/MAX anticipates the merger to close on August 24th, which it did. Kenvue received approval from the NZ Commerce Commission subject to divesting the Kenvue feminine hygiene business. And lastly, ATKR filed its transaction with the Austria Competition Authority. Oh, we're not done. You can see Warner Brothers rally on page five. You can see the biggest arb spreads. Warner Brothers is now number two from number one. We're long that from a 20% arb spread. It's now only 7%.

You can see there are links to all of these updates on page six.

## Navitas and Claros Acquisition - 22:00

But if we jump to Navitas, on page seven, that's another deal that was not picked up in the biggest arb spreads, but Navitas has agreed to acquire power management technology firm Claros in a deal valued at 232.8 million. The consideration includes 216 million of cash and stock at closing.

Founded in February of last year, Claros basically makes vertical power delivery and integrated voltage regulator, IVR, technology for next-generation AI data centers. Navitas said it would more than double its 2030 server serviceable addressable market to over 8 billion and accelerate revenue growth and margin expansion. The company said the technology addresses a major growing challenge for AI infrastructure — delivering thousands of amps to increasingly power-hungry processors at extremely high speed. Traditional voltage regulators distribute power across circuit boards, creating losses and limiting how much power can reach AI chips. Claros's technology moves the power conversion closer to the processor, reducing the distance power needs to travel and potentially improving efficiency and power density. This acquisition significantly expands their addressable market. So on pullbacks, this NVTS could be interesting.

## Arbutus Odd-Lot Tender Offer - 23:26

Now Arbutus is a special situation, something we've done before. It's basically an odd-lot tender. So in terms of the offer structure, Arbutus, or ABUS, which we're long from very low levels, like $2 a share, even less, before they won the lawsuit from Moderna — now obviously has loads of cash on the balance sheet, and what they've decided to do is buy back shares from the public.

So what they've done is they've launched a modified Dutch auction self-tender to buy back up to 230 million of common stock at a clearing price of between $5 and $5.75 per share. We found this from our new special sits screener, which should be coming out next month for you guys. And it basically has all of the tenders, all of the M&A deals, that you could screen for and do your own work on if you find them interesting. The purchase is funded in part by 178 million in net proceeds from the recent Moderna patent settlement we talked about, which succeeded.

Now, the odd-lot priority mechanism — the trade leverages this odd-lot priority provision. So shareholders owning 99 shares in every brokerage — so if you have 10 brokers, you can own 99 shares in each — who tender their entire position are exempt from proration. So if, for example, the company decides to buy back shares at 5.50 — it's between 5 and 5.75 — you would effectively make 42 cents, because it closed at 5.09 after hours on Friday. But if you did it at a close at 5.07, you make 43 cents per share at the 5.50, across, if you have 10 accounts, 990 shares.

In terms of the timeline, the offer expires on September 29th at 5:00 p.m. Eastern, so you actually have a full month to do the trade. In terms of the expected settlement or payment, it's supposed to happen on October 2nd. And the key downside risk is the final clearing price is unfixed, leaving exposure to a clearing price below the entry purchase cost if the auction clears at the lower bound of five. But at least you know what your downside is.

## Iran War & Sanctions Update - 25:37

In terms of the Iran war, we'll just go really briefly through this. On Monday, Scott Bessent issued explicit warnings as part of what he called an economic D-Day against Iran. Unfortunately, this has just not worked. The bombings continue. He basically said that the US is sanctioning more than 60 entities, individuals, and vessels. He said no one is above the reach. Targeting gray spaces — Bessent emphasized that the US would no longer tolerate entities operating in the gray spaces of the conflict, cautioning that any economic engagement with Iran's regime would expose third parties to the full reach of American power and exclusion from the US dollar system, which is quite bad for the US itself if it continues to push people out of the US dollar system. President Trump also actively made direct calls to foreign world leaders with respect to a request to cease trading and sever economic interactions with Iran.

And then the impact on China and other major partners. The reason why oil hasn't gone up much, as we saw last week, is that China is reducing imports again. One reason for that is China's economy is just weak, and we are anticipating a Chinese stimulus to come out. The other reason why this is happening is simply because China, I think, is managing oil prices. It had several hundred million barrels of oil in its reserves, much more than the US, and it was purposefully importing less to keep oil prices capped.

## Global Oil Inventories and US/Canada Trade War - 27:14

Now, you can see global oil inventories are set to reach historic lows if the strait doesn't reopen soon. So it's becoming more and more of a risk, especially after midterms if Trump decides to escalate the war again.

So the trade war is also continuing against Canada. Trump said he would place 50% tariffs on Canadian automobiles, trucks, and automotive parts starting in January. The threat came after Canadian Prime Minister Carney vowed to hit the US with retaliatory levies following the collapse of trade talks over last weekend. This is the latest escalation in a tit-for-tat trade conflict that could disrupt supply chains and threaten the future of the USMCA. Approval ratings have been falling because of all these tariffs and wars, which are resulting in inflation fear.

## US National Debt and Treasury Actions - 28:29

Now, earnings have been tremendous this year. What can derail them? Everyone's focused on Jackson Hole, which ended up being hawkish, but the real risk is geopolitical escalation after the November elections, which could force Warsh and crew to raise rates in December. Trump can't contain himself already with Canada, Iran — and imagine what comes after November. You can see our national debt reached over 40 trillion, an unbelievable number, last week.

We're going to have to accept some sort of a higher inflation if we monetize the debt, but that's a discussion for another time.

## Treasury Actions and Bond Market (Bessent) - 28:55

Bessent learned last week that he can't bring long-term rates much lower in the short term without blowing the capital in this TGA. Basically, Bessent last week sent the bond market reeling, albeit temporarily, with his plan to increase so-called buyback operations, whereby the US Treasury uses cash on hand to buy back some of its long-term notes and bonds from willing sellers. While the practice is intended to ensure ample market liquidity, especially in older off-the-run issues, the line between ensuring liquidity and influencing prices can blur easily. These operations, even if scaled up as Bessent described, are quite small relative to the total size of the Treasury market. A few billion dollars of purchases cannot meaningfully move bond market prices unless they signal a credible commitment to bring down long-term rates, which we haven't seen yet.

To that end, there were reports in the financial press that the Treasury could use nearly 1 trillion of its TGA account, but it would need permission, and this could actually be contested in court by the Supreme Court. Now, with the benefit of a few days of reflection, the bond market was not impressed by any of this Treasury buyback, as it correctly deemed these actions either ineffectual in the case of expanded buybacks or unlikely in the case of the TGA drawdown. Effectively, the market is saying it's going to be hard for you to upsize these bond purchases even more, which is what we need to bring yields down. So yields remain in the same range they have been over the past month, although the drop in WTI crude over the last week may have helped the 10-year come down slightly from its recent high. But what I would say is that what we are worried about is that the 10-year goes up again next week with oil prices rallying because of the escalation over the weekend.

## Relationship Between GDP, Inflation, and Interest Rates - 30:53

And again, I don't understand why we are doing all this escalation right before midterm elections.

So if you look at the graph at the bottom of page 11, higher oil prices in 2026 have meant higher inflation, which has translated into higher interest rates. But even geopolitics' impact on rates tends to be limited and short-term in nature. The true driver of rates over the long term is the performance of the economy, as we'll discuss in a minute.

If you look at nominal GDP versus 10-year yields, it turns out that nominal growth, which includes both real output and inflation, and nominal interest rates are closely related. Excluding the period surrounding the pandemic, nominal GDP growth has not been this high since the mid-2000s, which is also the last time the 10-year Treasury was this high. We do not view this as a coincidence. The chart above shows clearly that the 10-year rates have generally been lower than nominal GDP growth for most of the last 20 years. One reason for this could be that the Fed has, for much of this period, been an uneconomic buyer of government debt as it expanded its balance sheet to suppress yields. Another could be investors' collective belief in 2022 that short-term interest rates would remain close to zero in perpetuity and that high inflation would no longer be a problem.

This next chart suggests that both expectations are valid. Expectations of lower short-term rates fell after the 2008 financial crisis, as did the term premium, which is the extra yield that investors demand to hold long-term securities. You can see this graph on page 12.

The Fed's post-GFC strategy was to buy bonds and bring down the term premium while also keeping short-term rates fixed at zero, made possible by the low-growth and low-inflation economic backdrop. But in today's period of high fiscal deficits and above-target inflation, easy monetary policy is likely to push the 10-year higher rather than lower. Indeed, even the recent run of negative data surprises, with weak labor, etcetera, has failed to push down yields. You see the 10-year Treasury yield and economic surprises below that in a graph on the following page, page 13.

We think that we know why. Most of the bond data has been limited to housing and employment areas. With housing, the causal relationship with rates tends to point in the opposite direction. Higher rates lead to poor construction and sales, and weak hiring, while potentially a sign of an economic slowdown, may also signal that firms are able to produce more without needing more workers. In other words, their profit margins have widened.

We will talk about profit margins later in this call. The broad softness in the data has not stopped analysts from revising corporate earnings estimates higher or stirred a flight to safety out of equities into cash or bonds. This may be because categories of economic data that are more closely correlated with financial market behavior, including business surveys and capital goods orders, have been hot lately.

I'm just going to get some water in a second.

## The Fed's Stance and Jackson Hole Speech - 36:03

Okay, so the White House and the Treasury don't want these leading indicators to turn lower. But a Federal Reserve facing persistently high inflation and few acute concerns about a dramatically slowing labor market may feel different. The Treasury, of course, and the White House have delivered most of the market-moving policy news over the past few weeks. But this week, investors' calendars were blocked off, as they always are in late August, for the Fed Chair's remarks at Jackson Hole.

This year, the presence of a new Fed Chair made the event even more anticipated than usual. Chair Warsh's public approach in the opening months of his tenure has been to do little and say even less. In keeping with this practice, he provided characteristically few thoughts on how monetary policy might need to evolve in the near term to deal with above-target inflation. But he did close the speech with a frank assessment of inflation being too high and not improving meaningfully, and vaguely acknowledged that the Fed may need to react to it by saying, "We must be confident that the underlying inflation is moving to our objective, clearly and at a sufficient speed. Otherwise, we have work to do."

Prior to Warsh's Jackson Hole speech, the market-implied probability of a Fed hike for the September 16th meeting fell to its lowest level of the summer. His speech took those odds up a little, but as the graph shows below, the decision remains close to a 50% coin flip in the eyes of investors.

Employment and inflation data both came in soft for July, which may be sufficient to delay a rate hike for now. In addition, the rise in long-term interest rates, which Warsh has hinted he sees as the market doing the Fed's tightening work for it, have held at high levels. The one area of the economy that looks set to weaken is consumer spending, which we covered a few weeks ago. But tighter policy to bring inflation down could arguably be seen as helping address cost-of-living concerns.

Even with Warsh's remarks, saying what the Fed will do or should do in September remains a difficult call. In our description of what has brought interest rates back to cycle highs and what might be done to bring them back down, we've conspicuously avoided mentioning the federal debt or deficit. This is not because we're ignoring the effect fiscal profligacy is having on the bond market, it's simply an acknowledgement that there's currently no political will to tighten fiscal policy via tax increases or spending cuts or entitlement reform. Monetary policy and incoming economic data are likely to be the only two durable influences on the bond market for the time being, perhaps with the occasional geopolitical shock thrown in.

Short of an economic collapse, we don't expect bond yields to move dramatically lower in the near term, although yields could peak, which offers investors a window of opportunity to access some of the highest yields in a generation. Those who live in high tax areas should pay particular attention to the chart below, which shows the benefit of moving from taxable to municipal bonds using an apples-to-apples taxable equivalent yield comparison. Using your marginal tax rate, you can see that some muni bonds — for example, New York muni bonds — when you divide by your marginal tax rate, can provide yields that are much higher than corporate yields or US Treasury yields.

Yields on longer-dated municipal bonds in the 10 to 20-year range look even more attractive, but those bonds carry significant risk of a market loss if rates move much higher from here. One thing that ironically makes us comfortable with accepting more duration in municipal portfolios is that everyone thinks that interest rates will go higher. Investors who stake out futures positions in the 10-year are about as net short as they've ever been in history, right? And that's including people like Druckenmiller.

Longer-term investors who rely less on positioning data to make investment decisions can rest easy that relative valuations also point to the best entry point for longer-duration bonds versus stocks in nearly a generation. Our enthusiasm for bonds is by no means unbridled. Another string of hot inflation data, more likely given the rise of gasoline and diesel prices in August, could send rates even higher, particularly if the Fed is slow to react. And as we wrote above, none of the Treasury's gambits to influence the yield curve are likely to work even if they pass legal muster.

## Economic Update & PCE Data - 40:44

In terms of a quick economic update, the breakdown of the Canada-US trade talks triggered 50% US tariffs and Canadian retaliatory threats. Treasury yields whipsawed as heavier Republican and private borrowing pushed rates up, even as Treasury tried to counter it with doubling its buyback of bonds. US national debt crossed 40 trillion, having doubled in just a decade, with debt servicing costs up 15% this year and now 20% of US tax revenue. Oil stayed elevated on stalled US-Iran talks and has rallied a little bit now with strikes over this Sunday. On the labor and consumer, US labor data held solid with initial claims at 208k, beating consensus, and retail earnings gave a mixed read on the consumer, with a more cautious Lowe's against a more upbeat Target, and with Walmart missing same-store sales estimates by the most in years.

In terms of PCE, July PCE inflation, the preferred metric at the Fed, hit 3.7%, above expectations of 3.6%. Month over month was plus 0.2 versus estimates of 0.1. Core PCE was 3.3 versus expectations of 3.3, so in line, but the second highest reading since October 2024. Month over month was plus 0.2% versus an estimated 0.2%. The inflation front is calm; core inflation came in exactly as experts predicted with no bad surprises, meaning the Fed doesn't need to panic or raise interest rates. Growth is slowing — economic growth cooled down as expected, and consumer spending flattened out while inflation remains above the Fed's 2% target.

So why are futures weaker? Even though inflation wasn't worse than expected on the PCE day, it's still sticky because price pressures haven't gone away completely. Traders realized that the Fed isn't getting an immediate green light to aggressively slash interest rates, which led to some quick profit-taking in the futures market and a correction in crypto after a massive rally. You can see the core PCE monthly and the core PCE year-over-year in graphs on page 17, and then you can see the Bloomberg CPI core services less housing, super core, graph on page 18 — with the fact that I'm less concerned with PCE because more than half of the core PCE came from portfolio management and investment advice fees, reflecting equity price gains in the market.

You can see that graph specifically on portfolio management fees on page 19. And the best thing that Kevin Warsh could do at this point, to be able to lower rates later, would be to reaffirm the Fed's commitment to PCE until it hits 2%, and that's what he tried to do. Suggesting an index change last meeting was shocking and revealed low levels of competence, which caused market volatility. Note, Kevin is already famous for poor judgment by being hawkish into and even after the GFC.

## Gold Speculation and Market Positioning - 43:56

Gold speculators went all in before the Friday hawkish discussion. There's a record 22.2 billion surge in net gold futures bets, the biggest increase in over a decade. Over a three-week stretch, speculative buyers poured over 22 billion net value into gold futures. This rapid accumulation is the largest single build-up in gold positioning in over a decade.

And Bessent wasn't able to prevent that. In terms of the dual engine of the rally, the 22 billion shift was powered by both 13.6 billion in brand new long positions and 8.6 billion in short covering.

By the way, as I'm going through this, in the SSR Q&A tab, I will also upload a model for PayPal, which I should have uploaded. Now, PayPal is not incredibly interesting here, but I think it's interesting because through the 50 — we bought at 51, we sold at 54. But I do think if it were to sell off again in a weak market tape, which it frankly could do tomorrow, it would be interesting close to 50 again.

So we could always buy it back again.

Let me just upload this model here. Let me see how much space this is going to take up. And let me just go into the Q&A tab here. Q&A for SSR tab. Let's see. And I just uploaded it there. It's only a one megabyte file. Okay.

## Market Momentum and the Dollar - 45:50

So going back to our discussion — there's been a lot of momentum buying from CTAs in gold. Traders aggressively paid high premiums for gold call options, with JP Morgan even talking about it, to secure upside exposure, taking open interest to the 93rd percentile of the two-year range. The weakening dollar ahead of the dollar bounce on Friday was fueled by falling real yields and expectations of dovish central bank policy, which didn't happen.

## Post-Jackson Hole Fed Outlook - 46:31

After Jackson Hole, if you go to page 21, the September hike probability basically rose to 60-70%, and the December probability rose to 90%. Warsh's remarks brought relief to investors, urging a strong determination in dealing with price pressures. So he is focused on inflation at the moment, and I personally don't think he has a choice, because the FOMC has 12 voting members — he doesn't really have a choice at this point with at least two dissenting members and many others that I think are worried about inflation. I do think that this is why Bessent had to come up to the plate and buy back bonds.

## Analysis of Bessent's Bond Strategy - 47:16

And what Bessent is doing is not really QE from the Treasury's perspective. It's different. They're not printing money, but what they're doing is they're issuing bills at the very short duration and buying bonds, long duration, which is effectively increasing the interest rate risk at the Treasury. And forcing investors to take more risk as well.

## Warsh's Commentary and Stance on Inflation - 47:43

He basically said that this is my standard: we must be confident that underlying inflation is moving towards our objective, clearly and at a sufficient speed. Which means that he is listening to the other Fed members. It doesn't mean that he's going to have to hike, however, soon.

So Catalyst Funds basically said that Warsh was successful in re-establishing confidence. He came across as a member focused on inflation and bringing back the Fed's 2% target, which was different than his last speech.

So Warsh's speech this time was far clearer and more hawkish. Although Seema Shah basically said that maybe Warsh was so confident because he actually expects inflation data to come down — but I doubt he could know how soon the war settlement would resolve. It's just that real estate inflation and other categories are coming down.

## Market Reaction to Warsh's Remarks - 48:49

So after Warsh's remarks, traders boosted the odds of the Fed raising borrowing costs, with swaps showing traders see more than 50% chance of a hike in September, barely more than 50%. It moved a lot. I'm still not convinced, but we'll see.

## Detailed Notes on Warsh's Speech and Psyche - 49:06

So, notes on what Warsh actually said — they're at the bottom of page 21. Nothing too consequential there, but you can read through those notes. And then on page 22, just some more reflections on Warsh's psyche. In July, Warsh kind of hinted that we are waiting for a new inflation measure. He took that off the table for now by being strict about the PCE. That was probably to satisfy the committee, and that is what the market takes as hawkish. Warsh famously does not give forward guidance, but if he did, he would have just signaled something closer to a September hike. This is according to Dario Perkins.

I think that while the probability of September went up, when I reflected on it with some peers on Friday, I just don't think that a September or October hike really makes sense right ahead of midterm elections. Even though the war was started by us, I don't think that it makes sense for Warsh to hike with all this uncertainty around it, especially because inflation is roughly flat.

## The Trump Mandate and Market Perception - 50:21

Two, the one part of the Trump mandate that he is very explicit about delivering on is trying to communicate that the market should trade everything but the Fed. Not sure if that will be an easy task, but he is at least trying every single time. Trump has been very vocal about that as well. Good data should be good news for equities, rather than good data means rate hikes and sell everything.

## Further Analysis of Warsh's Motivations - 51:03

I tend to think that Warsh is buying himself time until the new committees and everything he's done are in place. Number three, Warsh didn't really deliver the productivity speech that everyone had hoped for, saying that productivity would bring inflation down through robotics and AI. But I think this is a big part of what he's thinking.

So if we do get a hike in September or October, it would only be to satisfy the groupthink at the Fed now, to be able to deliver on the productivity side later, once he has some committee approval. Warsh just made life worse for Bessent. One could argue that being hawkish brings long bond yields down, so it's not clear that he makes it worse. But from a liquidity standpoint, Bessent needs to be back more next week at the first opportunity. It would be fun if, for example, Bessent bought 10 billion worth of long bonds. He has stated at least 4 billion in the past, and I think at some point he's going to have to increase that 4 billion for the market to believe him.

## High Beta Stocks and Correlation - 52:16

You can see the high beta stocks — the negative correlation to low volatility stocks has never been like this in the past.

## Q2 Earnings Season Update - 52:25

Now in terms of the earnings season update, with over 90% of S&P 500 companies having reported thus far, the index is on pace for one of its best earnings seasons this century, as overall revenue growth is tracking over 15% year-over-year and adjusted EPS growth is tracking 31% year-over-year. This adjusted measure excludes recent one-time gains on private investments for several large tech companies, gains which, if included, inflate GAAP EPS growth year-over-year to 118.5% and need to be taken with a grain of salt, given how big Anthropic's mark-to-market was for Amazon and Google's P&L.

Now this level of adjusted earnings growth, roughly three times the long-term average for the S&P 500, is typically only seen in post-recessionary environments where companies saw their earnings drop and then bounce back strong. These often coincide with easing policy and economic recovery, making these results even more impressive in the context of the second quarter macro environment being so poor, where interest rates and elevated geopolitical risks were prevailing headwinds to consumption and profitability.

## Impact of Large-Cap Companies on Index Results - 53:45

Now, while the index-level results were influenced by some of the largest constituents, trends under the surface show that ex-Mag 7 earnings on average is 20.7%. The median S&P 500 company is on pace to deliver 14% EPS growth and 6% year-over-year revenue growth, both accelerating from last quarter. At the same time, earnings surprises remain pretty high. Nearly 90% of the companies have beaten bottom-line estimates by an average of 29%, while 75% of the companies have beaten top-line estimates by an average of 3%, nearly double the average revenue beat over the past 10 years.

Lastly, at the sector level, we see healthy contribution as well, with 10 of the 11 sectors on pace to post year-over-year EPS growth, and eight of the 11 on pace for double-digit year-over-year EPS growth, led by energy, communications, and discretionary. It's important to point out, however, that surging growth in some of the largest companies and segments is skewing index and sector-level results during the quarter. Excluding Amazon, for example, would reduce the consensus consumer discretionary EPS from a record 92% to just 7%. So imagine that. Excluding Amazon and Alphabet from the S&P 500 would reduce the overall earnings surprise rate from 29% to just 10% — a sizable haircut, but a hearty beat rate relative to history is needed. Lastly, when excluding major market segments like semiconductors and Mag 7, we see underlying earnings growth for the rest of the S&P still at 15%, healthy profit growth but far from the 30% adjusted that we see. You can see the graphical representations of these on pages 23 and 24.

But while recent revaluations on private investments for some of the largest tech companies have meaningfully boosted GAAP EPS, at this point the ratio of non-operating income relative to core operating income is elevated but far from extreme levels, which is good. I need to get some water one second.

## Profit Margin Analysis - 55:25

While recent revaluations of private investments for some of the largest tech companies have meaningfully boosted GAAP EPS, at this point the ratio of this non-operating income relative to core operating income is elevated, but far from extreme levels, averaging around 9% at the end of Q1 and likely heading into low double digits for Q2. In the 1980s, for example, non-operating income accounted for over 25% of corporate EPS, thanks to the healthy yield generated on cash at the time. And historically, this non-operating income has risen alongside a pickup in IPO activity and private capital raising. This is a metric that bears watching, as it has historically risen in later-cycle environments. And while it doesn't suggest an imminent end to the earnings cycle, non-operating income growth as a percentage of total income can lead to more volatile earnings in future quarters, especially if there's a slowdown.

Moving on to page 25, another positive data point this season has been overall profit margins, which surged once again in Q2 due to the combination of accelerating top line, embedded operating leverage, and disciplined expense management at many companies. Overall, profit margins for the S&P 500 are at a record 16.9%, expanding by over 400 bps over the past year to the highest level since 2009. While the majority of margin expansion has been concentrated amongst the largest companies in technology and semiconductors, in aggregate eight of 11 sectors are seeing margin expansion, and approximately 60% of companies within the S&P 500 have generated some level of margin expansion over the past year. These measures indicate that expanding profit margins are becoming more broad-based across the large-cap equity universe, a bullish signal, as historically rising profitability has been supportive of a higher valuation multiple rather than higher earnings.

## The Role of AI in Corporate Profitability - 56:22

Now, in recent years we've suggested that disciplined expense management would be a vital tool in allowing companies to deliver and exceed earnings growth expectations. While corporate belt-tightening has certainly been a contributor to this season's remarkable earnings growth, we are seeing signs that AI integration is having a positive effect on corporate profitability as well. When bifurcating the large-cap US equity universe into the companies that have specific AI use cases in their businesses and those that don't, we can see that those using AI have garnered a greater share of earnings growth, positive revisions, and margin expansion, which is interesting.

## AI CapEx as a Primary Driver of EPS Growth - 57:22

The most substantial driver of the S&P 500 EPS growth over the past year has been the remarkable growth in AI-related capex spending. That spending results in EPS growth. Investments that continue to exceed expectations in both magnitude and duration eventually have to stop. Estimates from Morgan Stanley research as of April 2026 suggest that AI capex accounted for nearly 70% of S&P earnings growth over the past three years, making the trend in capex growth and hyperscaler capex commentary an essential watch item each quarter.

Now, once again in Q2, we saw hyperscaler capex estimates exceed expectations, topping already lofty estimates by an average of 13%, while forward guidance generally indicated AI-related capex will continue to increase in the coming quarters. Now the issue here is that while earnings growth is going up, free cash flow growth is going down — and free cash flow, I think, is more important than earnings. The beneficiaries of this capex continue to garner the majority of EPS growth within the S&P 500, and this quarter the spread in year-over-year EPS growth for the AI infrastructure segment relative to the non-AI infrastructure segment is a record 40%.

While elevated AI capex continues to erode free cash flow for the hyperscalers, which could lead to continued debt and equity issuance for the group in the future, it does help fuel EPS growth expectations for many key industry groups like semis, cap goods, and hardware, given their outsized weights in the industry. Eventually, this spending will normalize, and it's possible that 2026 will be the year where AI capex growth peaks, as we are on track for over 700 billion in hyperscaler capex this year, an increase of 90% versus last year. As the spending growth normalizes — consensus expects 30% growth in 2027 — we should anticipate some moderation in the earnings revision momentum for the S&P 500 as well. In this scenario, fundamentals and durable growth are likely to take on a greater role in markets. We may also see wider dispersion, as companies that can continue to deliver on expectations will be rewarded.

Second quarter earnings season has certainly been impressive, evidenced by the magnitude and breadth of earnings growth, above-average beat rates, and healthy margin expansion within the index. While some of these metrics are inflated by one-time gains and non-operating activity, even excluding these items, we see a generally healthy environment for corporate profitability today, fueled by rising nominal GDP growth and continued operational efficiency at many companies. At the same time, we recognize that the current outsized EPS growth rates for the S&P 500 will be difficult to replicate going forward, and the index is likely over-earning today relative to historical levels.

## Hyperscaler CDS Spreads and Off-Balance-Sheet Arrangements - 58:32

If you look at the following page, hyperscaler CDS spreads hit all-time wides last week. Goldman Delta One is borrowed. Now these CDS spreads could calm down if they stop issuing more and more debt, but I don't think that that is the goal. Hyperscalers have 3.1 trillion of off-balance-sheet arrangements, not all of which are actually guaranteed. But if you look across Meta, Google, Microsoft, Amazon, Oracle, Nvidia, and Broadcom, you can see that there is a lot there that doesn't need to be on an individual's balance sheet. And I wouldn't be surprised if a lot of these people are converting assets from vested stock into precious metals.

## Monday Market Recap - 59:48

In terms of the Monday market recap, we will quickly blast through the most important events throughout the week. So number one, Treasury Secretary Scott Bessent talked about zero leakage with Iranian sanctions. Pelosi reportedly bought Bloom Energy and Intel at the end of July. Apple is reportedly testing DRAM from Chinese NAND companies, which is a big deal. SpaceX plans to build Nvidia-powered AI compute in orbit. The SEC reportedly sent subpoenas to Wall Street banks around Situational Awareness. Treasury officials say Scott Bessent could tap 950 billion of Treasury General Account firepower to buy back bonds. I'd be surprised to see that right now.

SoftBank plans to issue 6.3 billion of debt, basically to put more money, I would think, into OpenAI. Amazon is developing fully automated delivery stations through Project Tetromino, an early-stage AI and robotics effort focused on automating the final warehouse stop before packages reach drivers. MicroStrategy increased its US Treasury reserve to 5 billion, established an additional 1.6 billion of cash, strengthening investor conviction. Hedge funds posted their largest weekly sale of US equity since the week of Liberation Day in April 2025. Druckenmiller wrote a WSJ op-ed piece criticizing the Treasury's decision to double long-dated bond buybacks.

## Tuesday Market Recap - 1:05:26

On Tuesday, the US reportedly offered Iran a deal to halt the siege and lift sanctions in exchange for reopening the Strait of Hormuz and ending proxy attacks. Axios also reported that Rubio told several foreign officials that the US does not plan to launch new strikes on Iran now. Well, that was just proven incorrect. Gold physical ETFs saw 6.4 billion of inflows on Tuesday. Intuit — we talked about the miss. Trump said the US Navy has removed or detonated all mines from international waters. We have to see if that's true. Canada is responding to US tariffs with new tariffs of its own.

On Tuesday, Anthropic is expected to tell IPO investors its total addressable market exceeds 30 trillion, topping SpaceX's 28 trillion. OpenAI's data center head, Chris Malone, left the company last week. ClickHouse has surpassed 350 million in annual recurring revenue, up 40% since May. JP Morgan reiterated its overweight rating on SpaceX with a 240 target. I don't know how. On Tuesday, OpenAI says its new Broadcom-built Palomino AI chip outperforms the Nvidia GB300, which is interesting. Top 10 most active options were Nvidia, Tesla, Apple, SPCX, Intel, Amazon, Micron, AMD, Palantir, and SoFi. Raymond James raised its Nvidia price target to 352 from 330 and reiterated Strong Buy.

Basically, Raymond James said the firm's CPU opportunity is becoming more important, especially for agentic AI workflows, even though CPUs are only about 3% of sales activity. Raymond James expects CPU revenue to reach roughly 5% of total revenue by 2028. Nvidia is trading at less than 15 times 2027 earnings, below the S&P 500 at 18.6 times, which is interesting.

## Wednesday Market Recap and Nvidia Earnings - 1:07:58

So PCE was the biggest event on Wednesday. Nvidia earnings, the macro and political updates you can read about on page 33. And those of you who are more interested in Nvidia earnings can just jump to page 35. Nvidia basically delivered a double beat, and posted 60 billion in net income. There was weakness in the US offset by Asia. Its US revenue was 60.1 billion versus an estimate of 67 billion, yet total revenue of 96.2 billion beat consensus by almost 4 billion, so I don't think people necessarily care.

Hyperscaler revenues were 48.71 billion, up from 43 billion, which is very positive. Revenue guidance was 108 billion, from 104 billion. Revenue ramp has been genuinely absurd. But you can see with Nvidia's outlook, that was the crown jewel. It's expected to grow by 70% by 2028 — their fiscal year is a little bit different. They're calling it a supply-constrained market, which it has been for the past three years. This should be also good for Broadcom this week. Mizuho on Nvidia said that Nvidia's results were okay, but not enough. We'll revisit that. You can see Nvidia's price graph on the following page — Nvidia's price targets after Q2 earnings. Basically JP Morgan, Mizuho, Melius, Goldman Sachs, Bernstein, Citi all raised their price targets. The White House is also weighing a new round of chip tariffs, which had horrible timing after positive news around Nvidia. The Trump administration is considering broader semiconductor tariffs that would extend beyond chips to products like servers, laptops, and gaming consoles.

## Software Sector Earnings Review - 1:10:42

Now, in terms of these chip tariffs, they came in at such a bad time, after Nvidia earnings, that the stock index basically gave up most of its gains and ended up down 2.3%. In terms of the software earnings, Salesforce crushed earnings with Cloudforce. CrowdStrike was a cleaner beat — net new ARR of 333 million, up 51% year over year and a company record, with the Falcon Flex ARR more than doubling. Revenue came in at 1.47 billion versus 1.44.

Okta also crushed. Then CrowdStrike, which was up 20% — those shares jumped in late trading after the company projected revenue that exceeded analysts' estimates. So a summary of their 2027 year forecast with EPS at 9.21 versus 9.05. That's on page 39.

## Special Situation: PRTH Activist Investor Analysis - 1:11:41

On page 41, we talked about this PRTH. So PRTH is effectively majority owned by an individual CEO. And what's happening is investors that have gone activist hold more than 4% of the stock. They could get a majority of the minority. What they are doing is they're trying to get the CEO to raise his bid from $6, or roughly five and a half times cash flow, to something like $17 to $19. In reality, I think they'll get to 7 and a half to 8. But there's a lot of upside if I were to be right on this. EV to EBITDA is around six times forward, trailing P/E is around eight times, forward P/E is around four to five times. The company's quite cheap. It does have debt, like 950 of net debt, but that debt is covered by the EBITDA for now. Let me know what you think.

Activist investors against the CEO, Tom Priore, include Steamboat Capital Partners and Buckley Capital. They've combined own 4.1% of the business and they could increase it. So you can think about — both funds submitted letters to the PRTH Special Committee, and basically the board has a fiduciary duty to represent the minority shareholders. So the CEO's initial proposal was a non-binding offer made to a special committee of independent directors. Under Delaware law, if it's structured as an MFW dual-procedure framework requiring both a special committee and a majority-of-the-minority vote condition at its inception, the transaction grants deferential business judgment review and prevents controlling freeze-out lawsuits. If the bid lacks a strict majority-of-the-minority condition, the special committee remains the primary defense line to squeeze out a higher price before recommending the transaction. Given vocal public opposition from minority shareholders, the CEO will likely need to raise the bid to secure the special committee's approval and avoid post-closing fiduciary litigation.

In terms of key catalysts, the special committee process should happen at the end of Q3, Q4, because it should only take about six to nine months after reaching a definitive agreement. Then there's operational progress and tuck-ins, which with severance can cost quite a lot. And then the LBO valuation gap, same thing, but they have all the upside when they calculate it. Using the LBO valuation, Buckley Capital is basically saying that because the company generates so much levered cash flow that it should be valued at $17 to $19 per share. But the sell-off in Q2 is because the company saw margin compression, right? It's three businesses — merchant solutions, PRTH has treasury solutions, and then payables. The payables segment is not doing well and they had to cut guidance from 1.01 to 1.04 billion down to like 900-and-something million, and so their EBITDA guidance got cut. And that's why the stock was down so much, not for no reason. So there's severe margin compression in their B2B payables segment. SG&A also rose 21% to 16.8 million as they hired more people. Net leverage is around 3.8 times, quite manageable at the moment. In terms of Wall Street downgrades, KBW cut its price target on the company from 750 to 650, which is still above where it trades now.

## Special Situation: PayPal Earnings and Valuation - 1:15:48

Now, in terms of PayPal earnings, we shared a model. So Stripe was reportedly interested in PayPal. It said it is no longer interested after the board learned a higher bid. PayPal trades at a very low multiple, as a value stock, under eight times forward earnings. It has 7 billion of free cash flow on the market cap. So 7 billion — so it's a teen free cash flow yield. It trades at 8 to 9 times P/E; we think it should trade at 10 for various reasons. It does have a net cash position on its balance sheet. Not only that, it is losing market share to competitors like Square, but it's still generating so much cash that it can effectively buy itself in less than 10 years.

So we actually, after the deal fell apart, we bought some PayPal shares at 51.65, we sold them at 54.40. It closed in the $53 handle area; we'll buy back close to 50 again. This is one trade you could do several times. If you look at the 2Q earnings, they actually beat. So we're not really worried about it having terrible earnings; it's been on track. On page 46, we have all the equity upgrades and downgrades for the week, it's kind of neat. And then on page 49, we talk about Korea hiking and Korea macroeconomic indicators.

## Q&A Section - 1:17:17

So with that, let's go to Q&A for SSR very quickly.

## Relationship Between Nominal GDP and Bond Yields - 1:17:24

So Bitcoin Tina is asking: the current dollar GDP was 8%, so this is nominal for Q2. At 5.25% 30-year yields, very cheap against the 8% nominal. No, I think that 8% nominal is very, very temporary. In terms of actual growth, growth is slowing quite rapidly because the consumer isn't spending. So we're at 1.5% real, and I think that nominal will be coming down. So I don't think that 8% number means that nominal yields are going to 8%. I think that what's more likely is that growth is going to slow down. But could yields go a little bit higher? Could the 30-year go to 6%? It's very possible.

So if nominal has averaged the 6% range in the last eight quarters, and it's 8% for this coming quarter, I think if you look at both earnings estimates and GDP estimates, this quarter is supposed to be peak. So who knows if it's going to fall from 8% to 6% or 8% to 5% on a nominal basis or 8% to 7%, but it is going to be sequentially declining. And so I do think, on a marginal basis, if earnings growth is peaking this quarter for the year and GDP growth is peaking this quarter for the year, I'm less worried about rates going higher if growth is decelerating both from an earnings perspective and a GDP perspective going forward. Now, if GDP had averaged 6% for the last eight quarters, then yields should have been higher for those eight quarters as well. So what I'm focusing on is the marginal change. So that's a good question. Could yields go higher if this war continues? Yeah, absolutely. Why they would continue this war going into midterm elections with approval ratings the lowest this year doesn't make much logical sense. There should be either a moratorium or some sort of temporary resolution until midterms are over. But we'll see how that plays out.

## Concluding Remarks and Additional Resources - 1:19:54

Now, there are a couple of other attachments that I've shared with you guys. For example, this Merrill on the US balance sheet and the yen, which is quite interesting. If you go to page two, we talk about how the yen is massively undervalued. If you look at exhibit 1A, it shows the yen's foreign exchange value in purchasing power parity terms — how much the yen buys in Japan versus how much it can buy in the rest of the world. And based on purchasing power parity, as you can see, the yen is currently as cheap as it's ever been, even cheaper than in the 1970s when it was deliberately undervalued. And they're also saying that the US dollar is overvalued. The US is not helping Japan purely out of kindness, it stands to benefit as well. The other half of this realignment of policies is to strengthen the yen.

So the US supporting the yen, selling dollars to buy yen, is actually positive for both countries. And then on US debt levels, Ariana Chu basically says that yes, US government debt has surpassed 40 trillion. However, it's not all bad news. Debt held by the public is 32 trillion, which is 100% of GDP. I still think that's a very high number. It's the highest I've ever seen in my life. Now, one of the other attachments, this global market outlook by Standard Chartered, is quite good. They talk about equity reviews on page 14, gold and crude oil on page 15, fixed income views on page nine. So I recommend that you guys quickly go through that.

And in the Citi piece, they look at 10-year annualized returns over private equity and global equities, and they say private equity has been declining for several years now. We tend to agree with that. So with that, what I'll do is we'll upload the recording. Thank you guys for your time. We will be sending additional special situations like the PRTH or the odd-lot tender from Arbutus, or PayPal if it hits 50 again. And we are working on improving the screener. It now has valuation metrics along with a description of every special situation in real time from SEC EDGAR and descriptions of the business. The next thing we're looking at is value drivers — how to figure out which one of these names is interesting to buy or sell. So hopefully the screener is available at the end of September, early October, but we will keep you posted.