Some Thoughts on Berkshire Hathaway's Q2 2023 Performance
In the second quarter, Berkshire's elephant gun turned into a bazooka
Happy Monday and welcome to our new subscribers!
Special thanks, too, to those who recently became paid supporters! ❤️
When I previewed Berkshire Hathaway’s Q2 2023 earnings report last week, most of my questions revolved around Warren Buffett and Ajit Jain’s ominous-sounding comments from the annual meeting — in which the Berkshire honchos hinted at some serious headwinds ahead for the company’s operating businesses and GEICO.
So, when Berkshire released its 10-Q bright and early on Saturday morning, I must say that the results came as a most pleasant surprise.
Not because of the big headline number — $35.9 billion in net income — which is distorted by mark-to-market accounting. But, rather, because this collection of solid performances — with a few notable exceptions — puts Buffett and co. in an enviable position to attack the market in the future.
The Berkshire cash machine rolls on and on — shrugging off difficult conditions for some of its operating segments to keep the money pouring into Omaha.
In Q2 2023, Berkshire Hathaway’s elephant gun turned into a bazooka. Now, Buffett and Munger just need to find something to fire it at…
UNTANGLING EARNINGS: Berkshire Hathaway’s bottom-line number of $35.9 billion is quite misleading. And we can thank mark-to-market accounting for that.
This accounting mandate, which records unrealized investment “gains” and “losses” in net income, adds an unneeded extra layer of confusion to Berkshire’s actual quarterly performance.
For example: Apple, alone, accounted for over $26 billion of Berkshire’s Q2 net income. Even though Warren Buffett did not sell a single share of the Cupertino-based tech giant in the quarter and this “gain” exists only on paper. Almost 75% of Berkshire’s reported profit came via a glorified accounting quirk.
Of course, no one knows this better than Berkshire.
In an accompanying press release, CFO Marc Hamburg wrote in bold: “The amount of investment gains [or losses] in any given quarter is usually meaningless and delivers figures for net earnings per share that can be extremely misleading to investors who have little or no knowledge of accounting rules.” Amen.
In Q2 2023, Apple was jet fuel for Berkshire’s rocketing bottom line. But, in the future, it could just as easily become a massive anchor weighing down the company’s results. Again, all without selling a single share.
This is why Warren Buffett and Charlie Munger recommend that investors focus on operating earnings — instead of net income — when judging Berkshire’s performance.
And, on that front, the news is still pretty good. Berkshire racked up $10.0 billion in operating earnings during the second quarter — good for a 6.6% increase year over year. (That percentage gain gets even better if you factor out currency effects.)
Here are some of the highlights…
Earnings from insurance underwriting soared by 74% to $1.25 billion.
Manufacturing, Service, & Retailing — as a whole — recorded a 4.3% gain.
A good quarter for industrial products. In particular, Precision Castparts surged 31.5% on “higher demand for aerospace products” and “improving manufacturing and operating efficiencies”.
Not so much for building products and consumer products, though. “Lower revenues at Forest River and nearly all of our other consumer products operations [were] partially offset by the impact of the Jazwares acquisition, which contributed revenues of $378 million in the first six months of 2023.”
The Service segment grew earnings by 8.9%, while Retailing fell slightly.
BNSF Railway suffered a big drop of 24% on lower freight volumes and higher non-fuel operating costs.
Berkshire Hathaway Energy was flat.
BUYING BACK BERKSHIRE: No surprise here — Warren Buffett reined in share repurchases during the second quarter. After shelling out $4.4 billion in Q1 to reduce Berkshire Hathaway’s outstanding share count by 9,581 Class A equivalents, he scaled back to $1.3 billion of buybacks as Berkshire’s stock price rose steadily throughout the quarter.
Oddly enough, though, most of his repurchase activity occurred in June — when share prices were the highest. I’m not really sure what to make of that. Perhaps some other capital allocation opportunities fell by the wayside, leaving higher-priced buybacks as the best outlet for Berkshire’s money.
Here’s the month-by-month breakdown:
April: $151.5 million
May: $117.4 million
June: $1.03 billion 👀
Just to put Buffett’s unusual buyback cadence into perspective, he bought 177 Class A shares at an average price of $472,004.70 in April and 627 Class A shares at an average price of $506,476.84 in June. Likewise, Buffett snapped up nearly 10x more Class B shares in June than he did in April — despite the average price rising from $308.57 to $335.55 per share.
¯\_(ツ)_/¯
After picking up the pace in June, Buffett seemed to slow back down in July. Between July 1 and July 26, Berkshire repurchased 174 more Class A equivalents.
SELLING STOCK: For the third quarter in a row, Berkshire Hathaway was a net seller of stocks — with $7.9 billion more in securities sales than in purchases.
Over the last few weeks, there has been a lot of chatter in financial media over whether or not Berkshire would pare back its stake in Apple as its stock price soared and its valuation rose to a higher-than-normal level. Some guessed that Warren Buffett might bank a little profit from arguably his greatest investment of all time.
I never really bought into this speculation — Buffett not only talks about Apple as if it’s a quasi-permanent holding, but he’s quick to bring up his ill-advised sale of some AAPL 0.00%↑ in 2020 as a mistake that he’s not eager to repeat.
Still, it’s not exactly crazy to wonder whether Apple’s price has gotten out over its skis and might be in for some valuation-induced headwinds in the future.
Nevertheless, if my back-of-the-napkin math is correct, Berkshire did not sell any of its Apple shares in the second quarter.
We did learn one other thing: Berkshire appears to have unloaded another chunk of Chevron — making this the twelfth consecutive quarter that he has tinkered with Berkshire’s CVX 0.00%↑ position.
GALLOPIN’ GEICO: The auto insurer didn’t quite match its standout Q1 performance, but no one really expected that. Happily, GEICO did manage to keep its surprising resurgence on track — scoring a $514 million underwriting profit and an impressive 94.7% combined ratio for the quarter.
That makes it two profitable quarters in a row after six straight in the red.
GEICO’s current plan is simple: trade growth for profitability. As such, the insurer dramatically slashed advertising costs — leading to 2.7 million less policies-in-force over the past year — which has been balanced out by 16+% higher average premiums per auto policy.
At the halfway point of 2023, GEICO is sporting a combined ratio of 93.7% over those six months. Back in May, insurance chief Ajit Jain predicted that GEICO’s full-year combined ratio would come in “just south of 100%”. Maybe Jain’s words presage a difficult Q3 and Q4 for GEICO — or, maybe, the auto insurer’s fortunes are turning around faster than anyone expected.
Congratulations to Todd Combs, pressed into double duty here, for leading GEICO out of the wilderness and back to profitability!
(Shameless Self-Promotion Time: An annotated transcript of Combs’s recent interview with the NFM podcast will be this month’s bonus for paid supporters!)
FLOAT ON: Berkshire Hathaway’s insurance float keeps getting bigger and bigger, reaching $166 billion in Q2.
The asynchronous nature of insurance — a company receives premiums from customers, but doesn’t pay out anything until claims come due — creates what is known as “float”. This collect-now, pay-later model leaves insurers holding large sums of money that can be invested to create a new avenue of profitability.
Even better, if the insurer’s underwriting breaks even, this float is basically free. The company pays nothing for access to lots of investable money. That’s exactly what happened in the second quarter, when Berkshire recorded underwriting earnings of $1.25 billion.
From the 10-Q: “Our combined insurance operations generated pre-tax underwriting gains in the first six months of 2023 and, consequently, the average cost of float was negative.”
Just imagine showing up at a bank and asking to borrow $166 billion at a 0% interest rate. You’d probably get laughed out of the building. But that’s exactly what Berkshire’s insurance operations have managed to achieve.
If there’s a secret to Warren Buffett’s success, that’s it right there. He expertly builds up float and then uses it to drive greater and greater returns for Berkshire.
BUFFETT’S MONEY BIN: Berkshire Hathaway’s cash pile (can we still call it a “pile” when it’s this big?) is quickly approaching Scrooge McDuck proportions. Warren Buffett and co. now have $141.9 billion of cash on hand — even more if you count the $5.4 billion held by the company’s Railroad, Utilities, and Energy section.
That’s an eye-popping number, but Christopher Bloomstran points out that cash as a percentage of total assets is “in line with the past 25 years and below 2016-2021 levels”.
I’ll never complain about Warren Buffett and Charlie Munger having a $141+ billion cash pile with which to attack the market. That dry powder should come in handy as the economic outlook remains relatively uncertain — and opportunities could pop up at any time.
Plus, with interest rates now up over 5%, Berkshire earns billions of dollars each year just for parking this cash in short-term treasuries.
In effect, Buffett is being paid increasingly large sums of interest to stay patient and pick out just the right pitch to swing at — and that almost always means good things for Berkshire shareholders.
Excellent summary, Kevin. Thank you!
Gets crazy if you add float & cash on hand - the $20 billion BRK keeps for claims, the thats over $260 billion of dry powder. That is sovereign level stuff, which is why Buffett is more of a champion of the current system. Where else will a person worth $117 billion control $260 billion of purchase powering exist. It wouldn't.