Business Case: Precision Castparts – The Difficulty of Valuing a Business

How much should you pay for a good business?

It sounds like a simple question. It is not.

Precision Castparts is a particularly interesting case because Warren Buffett first appears to have been too optimistic about the company’s earning power, while only a few years later the value attributed to the same business had fallen dramatically. Today, with the aerospace industry in a very different position, Precision Castparts may once again be worth substantially more than Berkshire originally paid for it.

The case is a useful reminder of just how difficult it is to estimate the long-term value of a business.

In August 2015, Berkshire Hathaway agreed to acquire Precision Castparts, or PCC, for $235 per share. The transaction closed in January 2016. Berkshire paid approximately $32.7 billion for the equity, while the total transaction value, including net debt, was around $37.2 billion.

It was Berkshire Hathaway’s largest acquisition at the time.

And it was not cheap.

Based on PCC’s financial results around the time of the acquisition, Berkshire paid roughly 3.7 times revenue, 12.3 times trailing EBITDA and around 21 times trailing earnings. Buffett was well aware of that. At the time, he acknowledged that Berkshire was paying a very high multiple.

There were good reasons for paying a premium.

Precision Castparts was an exceptional industrial business. It manufactured highly engineered castings, forgings and fasteners used primarily in aircraft engines, airframes and power generation. These are not commodity products. They are technically demanding, often mission-critical and difficult to manufacture.

Suppliers need years of experience, certifications and close relationships with customers. In aerospace, switching a qualified supplier can be both costly and time-consuming.

PCC therefore had many of the characteristics Buffett traditionally looked for: a strong competitive position, difficult-to-replicate capabilities and exposure to industries with high barriers to entry.

The problem was not the quality of the business.

The problem was the price Berkshire paid relative to the earnings it expected the business to generate.

By 2020, that became clear.

The pandemic caused an unprecedented collapse in commercial aviation. Aircraft production fell, airlines postponed deliveries and PCC’s results deteriorated sharply. Berkshire eventually recorded approximately $9.8 billion of after-tax impairment charges related to Precision Castparts.

Figure 1. Warren Buffett on Precision Castparts: “I paid too much for the company”

Source: Berkshire Hathaway Inc. (https://www.berkshirehathaway.com/letters/2020ltr.pdf)

Buffett was unusually direct about what had happened. In his 2020 shareholder letter, he wrote that he had paid too much for the company because he had been too optimistic about PCC’s normalized profit potential.

That distinction matters.

He had not necessarily been wrong about the competitive position or the quality of Precision Castparts. He had been wrong about how much earnings that quality would ultimately produce and therefore about the price Berkshire should have paid for those earnings.

This is one of the most common mistakes in investing.

You can correctly identify a very good business and still make a poor investment if your assumptions about its future economics are too optimistic.

Then the environment changed again.

Commercial aviation recovered, aircraft production increased and demand for aerospace components strengthened. More importantly, the industry began facing shortages in precisely the type of highly specialized castings and forgings that PCC produces.

Capacity in these businesses cannot simply be added overnight. Manufacturing is complex, qualification periods are long and customers cannot easily replace suppliers. Assets that appeared deeply impaired during the pandemic suddenly became strategically scarce.

Precision Castparts’ results recovered substantially. In the first half of 2026, revenue reached around $6 billion, increasing more than 11% year over year and putting the company on a path toward approximately $12 billion of annual revenue.

Then, in September 2026, GE Aerospace agreed to acquire Consolidated Precision Products, one of PCC’s smaller competitors, for $11.75 billion.

The transaction is particularly interesting because GE is paying approximately 26 times CPP’s expected 2027 EBITDA before synergies. CPP is expected to generate around $2 billion of revenue in 2027.

Precision Castparts is several times larger.

Using the GE transaction as a reference point, some estimates now suggest that PCC itself could be worth close to $100 billion (https://www.cnbc.com/2026/09/12/buffetts-confidence-in-troubled-decade-old-acquisition-finally-pays-off.html).

That number should be treated with caution. It is an implied valuation, not an actual offer for Precision Castparts. GE is also a strategic buyer paying a very high multiple for an asset that can help solve an important supply constraint.

Still, the comparison is difficult to ignore.

Berkshire paid roughly $37 billion for PCC in 2016. Five years later, Berkshire wrote down a significant portion of the investment and Buffett concluded that he had materially overestimated the company’s earning potential. Another five years later, the economics of the industry look very different and comparable transactions may imply a value dramatically above both the pandemic-era assessment and Berkshire’s original purchase price.

It is tempting to say that Buffett was wrong twice.

He was clearly too optimistic about PCC’s normalized earnings when Berkshire bought the company. In hindsight, however, it is equally interesting to see how different the business looks only a few years after Berkshire concluded that a substantial part of its original value had been impaired.

The more important lesson is not whether Buffett was technically wrong for a second time.

It is that the range of possible outcomes for the same business proved far wider than anyone could reasonably have predicted.

When investors value a company, we usually build models around revenue growth, margins, free cash flow and discount rates. If the model is detailed enough, the final number can look remarkably precise.

The spreadsheet may be precise. The future is not.

In 2016, almost nobody was forecasting that a global pandemic would temporarily shut down much of commercial aviation. In 2020, almost nobody was forecasting that several years later aerospace supply constraints would become severe enough for a strategic buyer to pay 26 times forward EBITDA for a castings manufacturer.

Both developments had a major impact on the economics and perceived value of Precision Castparts.

Neither was realistically forecastable.

This is why I believe valuation should not primarily be about finding one supposedly correct number. Its purpose is to understand the range of possible outcomes, identify the assumptions that matter most and determine how much room we have to be wrong.

What if normalized margins are lower than we expect? What if the competitive environment becomes more favorable? What if an asset that looks ordinary today becomes scarce ten years from now?

We should try to answer these questions. But we should also be conscious of the limits of our ability to answer them.

The further we look into the future, the less we actually know.

Precision Castparts illustrates this unusually well. One of the greatest investors of all time correctly identified an exceptional business, but initially overestimated its earning power. A few years later, the outlook had deteriorated so much that Berkshire impaired billions of dollars of value. Today, the same business may again be worth far more than either of those assessments suggested.

That is not a criticism of Buffett.

It is a reminder of how difficult valuation really is.

A good investment process should therefore avoid false precision. We should try to own good businesses, pay sensible prices and build enough margin into our assumptions to survive outcomes we did not predict.

Good investing is not about knowing exactly what a business will be worth ten years from now.

It is about recognizing how little we know and making good decisions despite that uncertainty.