gfp

The theme for quite a while now has been 'long scarcity', short abundance. Software, outside of some incredibly compute intensive software like Claude, cannot be scarce by its nature. The stuff that makes it such an unbelievably good business - that incremental licenses incur almost no incremental cost, if I can make one copy I can make 1000 copies, etc is the same stuff that ensures it will never be scarce. Meanwhile, we have actual physical world bottlenecks all over the place. Starting with energy and moving all the way down the five layer cake of the AI buildout. Everything that is scarce is in the physical world, including certain commodities. Memory can be scarce, GPUs can be scarce, CPUs can be scarce, capacity in a TSM fab can be scarce, Grid capacity or generation turbines can be scarce, Helium can be scarce. Office 365 licenses are not scarce.

All sorts of funding shorts routinely get squeezed. You are seeing it recently in GFS and CRWV, which were (still are) popular funding shorts.

That is correct - I have previously asked the company and the TRS gains/losses are taxable each quarter. Conversely, trades in an issuers own common stock are not taxable to the issuer. There is not a built-up realized gain being realized when they retire a portion of the TRS position. The opportunity is to buy a block of shares from the counterparty who should hold that block of shares as their hedge to facilitate the TRS. Buying the hedge shares form the counterparty is not required but if you want the block, I"m sure the counterparty is happy to oblige

Berkshire isn't cheap. It's just slightly cheaper than it has been for a while and around the same price it traded at two years ago. Nobody is getting caught short on BRK stock so it isn't melting up with the market. Maybe Abel pulls an Allbirds at the AGM

I was hoping BRK would hit $999.99 Billion market cap today but we didn't get there. Maybe tomorrow

Greg / Berkshire is not buying. Berkshire stopped buying shares 2 weeks before March 31st. Which is coincidentally when the price started falling. A couple days after the AGM (the morning release of the 10Q) will put Berkshire back in an open period to repurchase their shares.

Everybody has generally been using this catch-all thread but there is an AJG thread

Congratulations Saluki, you are going to be a great Dad

As a stupid American I always have a hard time remembering how to figure crore. This is approximately $108 million USA bucks? Am I doing that right?

I assume you are referring to automotive sales declines in China or some other business headwinds but I thought it was a funny reference because of their recent video showing their vehicle using its suspension to literally jump over potholes (and road spikes - take that state troopers...) https://www.youtube.com/watch?v=hW6lFowulFw

please excuse the blasphemy of including a price chart but I've been surprised by BWIN's outperformance since the general broker group put in their most recent attempt at a bottom. It couldn't even fill its gap after the earnings pop retracement. Perhaps the strength is due to Baldwin being singled out as more exposed to disintermediation risk because of its perceived exposure to "low complexity" segments (main-street commercial, personal lines, etc) and this "strength" is actually short exposure being caught offsides and covering. Could be that their perceived weakness around AI disruption risk ended up causing short term outperformance.

Historically Fairfax has been too small and the shares too illiquid to be worth investing in for Berkshire. Later when the new investment managers joined BRK and took smaller positions, FFH never showed up - but could technically have been hidden in Berkshire's disclosures - FRFHF would not show up on 13Fs as an OTC stock and FFH wouldn't show up as a foreign stock. Fairfax has never shown up in any of Berkshire's NAIC filings but there are ways for Berkshire to hide the identity of individual investments on those filings as well (the prime example is buying shares inside "Harney Investment Trust" and similar entities formed for this purpose). Ted or Todd could have owned an undisclosed position inside the pension funds they managed for Berkshire and it would have remained undisclosed. I doubt it was in there but you can't prove it either way. There are no 10% economic holders of Fairfax. Even Prem, with voting control, only owns a 6.2% economic interest in FFH worth about $2.4 billion US dollars. Baillie Gifford was the other large disclosed shareholder of Fairfax but had been reducing the size of their position which seems to have peaked in 2022 and my understanding is that are now below 5% and no longer need to disclose. I would assume OMERS owns FFH shares but they don't own enough to disclose. They are bigger holders of FIH percentage-wise but probably not in dollar terms. Then you've got passive funds and ETFs like Vanguard Total International Stock Fund, a disclosed position by Fidelity Contrafund that is probably in the $550m range (Contrafund / Danoff were decently big sellers of FFH in 2025). Long story short, there are no large owners of Fairfax and Berkshire would have a hell of a time building a position that could move the needle even if they wanted to. I would not be surprised in the least of Warren had a small personal holding to "subscribe" to the annual report and followed the company for decades. When Berkshire (I assume WEB) started buying Chubb stock they bought their first tranche at an average cost of $218.46 per share for an investment of $5.9 Billion and have added to that several times. CB was just a much larger and more liquid stock. (for those curious, that initial batch of CB stock purchased for $5.9 Billion is currently worth $8.934 Billion and the total CB investment at last disclose is now worth $11.32 Billion)

Gift link to WSJ article on Abel’s first 100 days https://www.wsj.com/finance/berkshire-hathaway-ceo-greg-abel-first-100-days-a42fcf27?st=LP6W5K&reflink=desktopwebshare_permalink

I tried out Opus 4.7 for the first time by asking it to speculate on the deal size of CSU's Derbysoft acquisition -------- I'll search for information about this acquisition to give you the most accurate estimate.The deal terms are confidential — the companies did not disclose financial terms of the transaction. So any estimate requires triangulating from what we know about Derbysoft and Constellation's typical acquisition pattern. Let me pull some data to build that estimate.Now I have enough to triangulate. Let me also check Constellation's typical acquisition profile.I have enough to build a credible estimate. Here's the triangulation: Bottom line: likely ~$75M–$175M for the majority stake, with a central estimate around $100M–$130M What we know Deal terms were not disclosed. Constellation rarely discloses sizes on individual acquisitions unless they are material (typically >$100M), which itself is a useful signal — this one likely falls near or just under that threshold. Target profile (Derbysoft) Revenue: Third-party estimates place Derbysoft at roughly $55M–$65M annually, with one source citing ~$61M and ~330 employees at ~$184K revenue/employee. Another trade-data source lists a broader "up to $500M" range, but that appears to be a category bucket, not an actual figure. Business: Hotel distribution connectivity (CRS/PMS-level), digital marketing, content management, and newer AI/flight modules. Classic vertical market software — exactly Constellation's strike zone. History: Founded 2002, only ~$15–30M raised historically, profitable enough to have never needed to IPO. Recently made its own tuck-in acquisitions (Arise, PKFARE), suggesting reasonable cash generation. Constellation's acquisition math Constellation targets ~25%+ IRR hurdle rates, which in practice has meant paying roughly 1–3x revenue or 5–10x EBITDA for smaller VMS deals (Juniper Group specializes in smaller/mid-sized tuck-ins within Vela). Larger/distressed deals like Optimal Blue have gone at ~7x EBITDA; smaller proprietary deals often go higher on revenue multiples but are still disciplined. Estimate build Applying a typical Constellation range to ~$60M of revenue: Low case (~1.2x revenue): ~$70M enterprise value × ~60% majority stake ≈ $45M Base case (~2x revenue): ~$120M EV × ~70% stake ≈ $85M High case (~3x revenue, reflecting travel tech premium and AI/flight expansion): ~$180M EV × ~70% stake ≈ $125M So a reasonable range for Juniper's check is roughly $75M–$150M, with ~$100M–$130M as the most defensible central estimate for the majority-stake purchase price. The enterprise value of the whole business is probably in the $120M–$200M range. Caveats Travel tech commands a premium vs. Constellation's usual verticals, which could push the multiple higher than typical. Management retaining a minority stake is structurally important — Constellation often uses this to lower the upfront check and align incentives, which tends to compress disclosed headline numbers. The fact that Cantor Fitzgerald ran a process suggests Derbysoft likely attracted multiple bidders, which pressures the multiple upward. This is an estimate, not a leak. The actual number could land meaningfully outside this range; Constellation will only reveal it indirectly (if at all) via segment disclosures in future quarterlies.

Charles predictably pumping NextNav at the AGM! I figured it was up this morning because he was out there working the floor giving stock tips

Prem's exclamation marks getting a shout-out at the AGM - "the only thing Rick Salzburg couldn't take out - my exclamation marks"

Looks like Flow alkaline water in tetra-pak

I was just reporting what was said at the AGM. It was $27 million USD equivalent. Every million counts!

SPY pennies from an all time high and boring old BRK at the bottom end of its recent range. I assume they can start repurchasing shares again a few days after the next 10Q comes out. I don't believe they have been buying any of their own stock since March 17th or so

on Sanmar exit - it would have needed more capital, we didn't want to put in more capital, the controlling shareholder wanted to buy it, we thought we had other opportunities (not sure if this is past tense on purpose)

Ben giggles to himself.. "I don't think getting capital is going to be a problem for us" Spoken like a true son of a billionaire

Asheef on the mic asking about dividend capacity - Hari, why aren't you sending us cash bro?? Gobinath up next?

ha - it does seem like some of those Skyh pitches are written with a bunch of ChatGPT help and little editing to ChatGPT's style doesn't it? And since the LLMs use Reddit as a popular source there is some circularity to it all. I don't think any of that will change the outcome for Sky's business model though. Just a question of what you want to pay for it

As a resident of the US Gulf Coast I can confirm that refined products are abundant and considerably cheaper than the rest of the country. Our consolation prize for living in 'cancer alley' ...

https://www.cnbc.com/2026/04/15/allbirds-bird-stock-shoes-ai.html