The $11.75 billion customer

A jet engine missing one essential part is a very expensive paperweight.

Which makes the company supplying that last part rather interesting.

GE has agreed to pay $11.75 billion for longtime supplier CPP. Some of the most interesting potential profit from the deal wouldn't show up in CPP's accounts. It would show up in GE's.

That's a different acquisition pitch: don't just buy the earnings. Buy the ability to earn more somewhere else.

Also this week: Miro sellers swap one kind of expertise for another, and a quarter of Porsche's supercar-stake proceeds heads somewhere decidedly less glamorous. Pensions.

The week's transactions, with the price labels left on. $ means US dollars.

Imagine a supplier considering another production line. It counts the extra parts it could sell, subtracts the costs and decides the return isn't good enough.

Its customer runs a different calculation. Those parts could help finish more engines. The same factory investment suddenly looks much better.

Nobody has to be bad at their job. They're counting different profits.

That's the ownership hypothesis worth testing in GE's agreement to buy CPP from Warburg Pincus and Berkshire Partners. GE, a customer for more than 15 years, says it wants more casting capacity and closer coordination between engine design and manufacturing. Closing is expected in the second half of 2027. GE announcement

Under one owner, a production line can be judged on both the parts it sells and the engine production it enables. The investment budget follows a different calculation.

For a PE owner, that changes the exit preparation. Don't stop at showing how efficiently your factory runs. Show what a better delivery rate makes possible in the customer's factory, and which investments unlock it.

Then comes the awkward negotiation. Creating value for a customer doesn't automatically give you the power to charge for it. If only one buyer can capture the upside, you may have a compelling strategic story and very little competition for the business.

Nor does GE need to buy every supplier whose expansion it would welcome. A long-term order or jointly funded production line might achieve the same result for less. The acquisition earns its premium only where ownership improves on those alternatives.

There is a useful warning from another casting deal. In his 2020 letter, Warren Buffett revisited Berkshire Hathaway's purchase of Precision Castparts: “I paid too much for the company.” His mistake was overestimating normal earnings—not discovering that castings didn't matter. Shareholder letter, page 4

Precision Castparts is a different company; Berkshire Hathaway is not CPP seller Berkshire Partners. The relevant lesson is how easily confidence in an essential product can become confidence in an earnings forecast.

“They need us” starts the argument. “Here's what ownership lets them do differently” makes the case.

Selling a company can make you rich without making you qualified to invest the proceeds.

Especially when the next investment arrives in the same closing documents.

Bending Spoons has agreed to buy Miro, which it says has roughly $600 million in annual recurring revenue. Certain Miro shareholders will invest $295 million of their proceeds in newly issued Bending Spoons shares. Buyer announcement

That can be a sensible trade: less dependence on one product, exposure to a larger portfolio, and a share of whatever the new owner improves.

But the research job changes. Knowing why customers renew Miro isn't the same as knowing what Bending Spoons should pay for its next acquisition. Product judgment gives way to capital-allocation judgment.

And owning more products doesn't diversify every risk. If one owner's operating playbook is applied across the portfolio, a weakness in that playbook can affect several businesses at once. More logos; still one set of decisions at the top.

The useful diligence is therefore less “Do I like these products?” and more “Where do the returns come from?” Separate improvements in businesses already owned from growth purchased through the next acquisition. Then test what happens if attractive acquisitions become scarce or expensive.

The psychological trap is treating a buyer's enthusiasm for your company as evidence of its skill buying everyone else's. Neither follows from the other.

Accepting the offer and buying the acquirer deserve separate investment memos. Even when they happen on the same day.

Porsche completed its Bugatti Rimac and Rimac Group stake sale. Of approximately €1 billion in proceeds, €250 million will go toward pension funding. Porsche

Horsepower, heritage, pension obligations. Not the usual brochure.

There is a capital-allocation lesson hiding in the pension contribution: the most exciting asset isn't necessarily competing with another exciting asset. It's competing with every other claim on the owner's money.

That changes how to assess a disposal. “Would we like to own this?” is incomplete. Ask what keeping it prevents you from funding. A buyer with different obligations can reach a different answer without either side being foolish.

Porsche also lifted its automotive net cash flow margin outlook from 3–5% to 5.5–7.5%, reflecting the sale and pension funding. Both ends rose 2.5 percentage points. That's a substantial change in the cash outlook without evidence, from this transaction, that making cars became more profitable. Porsche's forecast bridge

Selling an investment can fund the next move. It can't become next year's operating plan.

Blackstone buys the components, not the chatbot. Flow Control Holdings supplies data-center liquid cooling as well as food, beverage and pharmaceutical markets. Management describes four years of investment in cooling and ten acquisitions under Audax. The interesting portfolio exercise is capability transfer: what can an existing industrial business credibly sell into a new market? The proof is qualified products and paying customers, not adding “AI” to a slide. Blackstone

Copart doesn't need more accounts. It needs the next bidder. ACV brings dealer-to-dealer digital auctions; Copart brings a global buyer network and more than 250 physical locations. The opportunity is matching a vehicle with a buyer who values it more—not merely combining user totals. A larger network adds little if the same dealers bid on the same cars. The integration scorecard should track incremental bidders, completed sales and net proceeds to sellers. Account growth is the easy screenshot. Joint announcement

Goodwin: you can buy the division without buying everything that makes it work. The Cerberus agreement excludes the refractory and technology divisions and requires an internal reorganisation. Shared services create the diligence trap: yesterday's allocation of the parent's IT or finance costs may bear little resemblance to tomorrow's cost of providing them independently. Start with the services the business needs on day one, price them, then reconcile that bill to the reported earnings. Original RNS

Tenet turns the financing relationship into an asset. Cerberus formed Tenet with management in 2021 to provide net-lease and sale-leaseback capital. CBRE IM's acquisition raises a different durability test: can the platform keep originating attractive transactions after the ownership changes? A portfolio of leases and a repeatable source of the next good lease are not the same asset. Cerberus

A1's growth story includes following up with homeowners who called but never booked. Before buying more demand, recover the demand already paid for. Read the A1 Garage Deal File.

Not another logo wall. A buyer-specific investment case with three lines:

Keep that upside separate from standalone earnings. And ask how many buyers can actually capture it. One enthusiastic customer is not the same thing as a competitive auction.

Editorial analysis, not investment advice.