Post New

I guess my short view:

Investments doing well, but operating subsidiaries are still weak.

Earnings in operating subsidiaries are better than they were, but still struggling. Valuations that use a multiple of those earnings as the non-investment leg are giving me much lower valuations than a multiple of book, with multiples chosen to give about the same numbers several years ago.

For example, Manufacturing/Service/Retail rolling-four-quarters real earnings peaked around the first half of 2022 and are down about 12% since then. That's not a one-off quirk, it has been a continuous gentle slide. Rails down 10% in the same time, but everybody knows that. A true optimist could see that as merely comparing current figures to a prior coal shipment bubble that we didn't appreciate at the time, but it's still down. Utilities are now merely weak rather than terrible.

Comparing the 2025 figure to three years earlier (actually the average 2-4 years earlier to avoid the quirks of any single number):
Inflation-adjusted book per share up inflation + 9.6%/year.
My inflation-adjusted value metric using a multiple of earnings for non-investments, up inflation + 5.2%/year.

That latter figure includes a cyclical adjustment upwards for Berkshire's share of KHC earnings. No impairment, and filling in the missing quarter.

Jim

In case anybody is interested, here are my recent quarterly figures from my "two and a half column" method. These are not intended to be the value of a share, merely some metric that rises at the same rate as the value of a share. Think of it as "true value times a single unknown constant". Inflation adjusted.

2020 Q4   519623
2021 Q1   549008
2021 Q2   590766
2021 Q3   610437
2021 Q4   611747
2022 Q1   618494
2022 Q2   591307
2022 Q3   575216
2022 Q4   598024
2023 Q1   572430
2023 Q2   589515
2023 Q3   562018
2023 Q4   592267
2024 Q1   604410
2024 Q2   616019
2024 Q3   639003
2024 Q4   645515
2025 Q1   646639
2025 Q2   664528
2025 Q3   685793
2025 Q4   687697

Quick summary of the method:
* Investments and cash per share at market value, but each stock holding capped to 21 times look-through earnings. Mainly, the Apple position gets tracked based on earnings progress, not price progress.
* Add 15 times net income per share for "steady things": operating subs, rails, utilities, equity method holdings, and a cyclically adjusted estimate of underwriting profit.
* Subtract 30% of float per share as a rough estimate of the overall long run drag on portfolio returns from always having to hold a lot of cash. Not the same as assuming 30% of float will be in cash.
* The occasional adjustment. For example, this year I put in an estimate of a conservative but "normal" year of earnings for KHC, as 2025 had only three quarters of income due to the new reporting lag.

Jim

* Subtract 30% of float per share as a rough estimate of the overall long run drag on portfolio returns from always having to hold a lot of cash. Not the same as assuming 30% of float will be in cash.


Just trying to understand what you mean by this, 2 questions:

(1) Why subtract any of it, since cash is not really cash? Substantially all the 'cash' is actually in short-term treasuries earnings a few percent, so why not just count it as an asset at face value?

(2) Of course any insurance company is going to have all or most of its float invested in fairly short-term fixed income securities, and this represents a drag on returns as opposed to a holding company that has all its capital invested in companies or equities. But if you have a $100 loan that you never have to pay back but can invest at 3%/year in the meantime, isn't it worth roughly $100?

Regares, dtb

Why subtract any of it, since cash is not really cash?...
(2) Of course any insurance company is going to have all or most of its float invested in fairly short-term fixed income securities...

The answer to the first bit is simply the second bit, nothing fancy.

My thinking was like this: an average portfolio is worth its face value, because most of the time most of the assets have a decent average earnings yield. It's worth what it earns. A portfolio with its hands partly tied, with forever some average "extra" allocation to cash earning on average nothing after tax and inflation, is worth less than its full face value because of that.

It sounds a bit harsh, but remember that my method doesn't subtract debt or the float liability!

Another way to look at it: It's a stake in the ground between the extremes of saying float is never a liability at all so all investments in effect belong to shareholders, and saying that the bookkeeping reality is 100% true and shareholders don't have any value interest in the assets held against that liability. Both extremes seem implausible, so I picked a 70/30 split. Investable float isn't worth zero, it isn't worth 100% as much as completely unfettered assets, so I estimated that it's worth maybe 70% as much.

Jim

Buffett has repeatedly said that insurance float is better than equity capital — if it’s cost free or profitable. Which it is now and has been for decades.

“Our float is money we hold but don’t own.”

And more importantly:

“If float costs us nothing — or less than nothing — it is better than equity”.

I’ve ballparked float value at 1.3X equity. I’ve seen estimates at 1.5x. Charlie seemed comfortable with 1.3 and said it’s rationally obviously 1.1x or higher.

The secret sauce of Berkshire has always been—its balance sheet carries huge liabilities more valuable than assets. Counterintuitive but true.

But you MUST HAVE supreme Underwriting discipline for that to be.. Which we have. And …must continue.

Buffett has repeatedly said that insurance float is better than equity capital......
I’ve ballparked float value at 1.3X equity.

IMO your understanding of float is flawed. When Buffett said float is better than equity, he meant that Berkshire would have to sell additional shares to raise equivalent amount of equity as float which makes such an alternative far worse for shareholders. He didn't mean that you could just take float and slap a multiple > 1 and simply add it to shareholder equity on the balance sheet.

I obviously don’t slap number on the balance sheet. I slap on the number in estimating valuation.

I obviously don’t slap number on the balance sheet. I slap on the number in estimating valuation.

True, but your understanding of float is still flawed. You can't just slap a multiple > 1 for the reason outlined.

Charlie Munger has advised we must assign a premium to the value of float , I always have, and assume as a disciple you should too. I would never have held Berkshire this long had I not. I would have estimated lower intrinsic value otherwise.

Equity has a required return. Float may not.

Shareholders expect 8–10% (or more) over time.
Policyholders do not expect an investment return — they expect claims to be paid.

If underwriting breaks even, float costs 0%.
If underwriting is profitable, float has a negative cost.

So if you can invest at 8%:
• Equity earns 8% but must justify 8–10%.
• Float earns 8% and only needs to justify 0%.

That difference — the spread — is economic value.

equity earning 8% isn’t automatically value creation.


If equity requires 9–10%, then 8% is actually destruction of value relative to opportunity cost.

Float earning 8% with a 0% cost?

That’s pure spread.

That’s the entire engine of Berkshire’s compounding story.

I know this is High School economics to many. But I suspect many if not most of our “partners” don’t understand this. We can disagree over the ACTUAL value. But it’s added value make no mistake about it.

@LongTermBRK

Of course float has value, and everyone agrees with that. What I disagreed with is your estimate of value of the float. Float is a liability since it's money owed to policy holders. Thus it is thus almost always invested conservatively by Warren & Greg. If the underwriting is done at a break-even combined ratio of 100%, resulting earnings are kept by the shareholders. Currently T-bills produce 3.5% pre-tax return, no where near the 8% return you presume (I take it after-tax). Assuming float grows at 3%, tax-rate of 20%, and a 10% discount rate, float is worth 0.4X. This is my base case valuation float.

If on the other hand u/w is done (LT) at a 97% combined ratio, float is worth 0.74X. This is the optimistic scenario.

So in my mind, float is worth somewhere between 0.4-0.75X depending on one's assumptions.

Instead of backing up your assertions with facts, you seem to unfortunately resort to unproductive rhetoric.

insurance float is better than equity capital — if it’s cost free or profitable. Which it is now and has been for decades.

Actually I recall a lot more comments to the effect that it's worth a lot, very much more than its booked value which is zero. But "better than equity" makes no sense to me at all, personally. Equity has no liquidity restrictions or time horizon or regulators.

As for profitable underwriting, I count that separately as a multiple of cyclically adjusted underwriting profit, the value of the business activities that give rise to the float. I use a very conservative estimated negative cost of float because an insurance business can have the occasional *really* bad year, even though supercat is a smaller piece of the pie these days.

Jim

I’m not going to respond to condescending insults—so let’s just see if I can explain in a way that even YOU can understand:

You’re correct that float is a liability in accounting terms. No one disputes that.

The issue is economic value.

If float costs 0% (or less), and can be invested at a positive rate, the value of float is the present value of the spread between investment return and float cost.

That is not rhetoric — it’s finance.

\text{Value of Float} = \frac{(r - c)F}{\text{discount rate}}

If:
• r > c
• float is durable
• underwriting discipline is maintained

Then float has positive economic value above book.

This is not my invention.

CHARLIE MUNGER said float is worth “more than 1.0×” and rationally “1.1× or more” if it is low-cost and durable. You might want to revisit your moniker if you seek to attack Charlie and my thinking on literally THE most fundamental matter in understanding what makes Berkshire uniquely valuable.

Warren Buffett has repeatedly stated that costless float is BETTER THAN EQUITY..

That is because equity carries a required return. Float does not.

If you assume:
• T-bill returns forever,
• 10% discount rates,
• and low reinvestment spreads,

you will, of course, get a low valuation.

But Berkshire does not invest float at T-bill rates over full cycles.

Historically, float has funded operating businesses and equities compounding at far higher long-term rates.

Your model implicitly assumes:
• Low returns
• High discount rate
• Limited reinvestment opportunity

That is a different business than Berkshire has been for 50+ years.

We can disagree on assumptions.

But the principle that low-cost, durable float has positive economic value above book is not controversial. It is the core of Berkshire’s model.

You can continue to argue float has minuscule value relative to equity as you outlined ..but..

Warren Buffett explicitly disagrees you.

Charlie Munger explicitly disagrees with you.

Attacking ME won’t make you right, and Warren, Charlie, and me wrong.

I've always thought of float as an asset but with a HUGE 72 point asterisk next to it.

In my personal balance sheet there's a reason many assets are in the "other" category (non cash, non marketable securities) - these are hard to value, illiquid, or possessing of any number of other conditions making them complicated.

Gemini mentions the 1998 meeting..a meeting I attended, where Buffett explicitly disagreed with you. My friend TexIrish and I along with a few other shareholders discussed this matter for many many hours. How much MORE valuable IS float? NO ONE argued less than 1.1x. The smartest Berkshire investor either of us ever knew, now deceased, argued the value was higher than either of us believed. Higher than I am now arguing.



Warren Buffett famously stated that "float has had a greater value to Berkshire than an equal amount of equity" during the 1998 Berkshire Hathaway Annual Meeting.

He elaborated on this "accounting irony" by explaining that while float is technically a liability on the balance sheet, its real economic value is superior to equity because it often comes at a negative cost—essentially meaning Berkshire is paid to hold and invest other people's money.

The Quote: "Float has had a greater value to Berkshire than an equal amount of equity... This statement would not be made if the float cost a couple percent a year, even though it would still be desirable. The current float is better than that, having a cost of less than zero with a profit attached".
The Reasoning: Replacing float with an equal amount of equity would technically increase the company's net worth, but Buffett noted it would result in less earnings for the company the following year because the float is generated through profitable underwriting.
The "Rocket Fuel" Comparison: In his 2021 Shareholder Letter, Buffett reiterated this sentiment, noting that although float is a liability, it has been "an extraordinary asset" that has grown 8,000-fold since 1967.

Key Distinctions
Cost of Capital: Buffett and Munger view float as the "best form of financing" because it is often interest-free or even profitable, whereas traditional equity or debt always carries a cost.
Economic vs. Accounting Value: On the Berkshire Hathaway balance sheet, float is recorded as a liability. However, because it is "long-enduring" and "costless," its true economic value is far higher than its accounting representation.

But Berkshire does not invest float at T-bill rates over full cycles.

Historically, float has funded operating businesses and equities compounding at far higher long-term rates.

It's been a long time since anything more than a single-digit percentage of the float amount has been in equities.
Historically, sure, it was the engine that drove Berkshire. Hasn't been for a long time. Maybe it will again? Hope so.
How to value it? How does the market value it?

Numbers I have to hand:
(Cash+Fixed)/Float

2010	105%
2011	92%
2012	100%
2013	94%
2014	101%
2015	99%
2016	102%
2017	109%
2018	105%
2019	111%
2020	113%
2021	109%
2022	92%
2023	111%
2024	195%
2025	220%

So in my mind, float is worth somewhere between 0.4-0.75X depending on one's assumptions.


Seems reasonable.

For those who disagree, you may just be talking at cross purposes: there is a difference between valuing the assets provided by the float, valuing those assets in combination with the offsetting float liability and restrictions, and valuing those two along with the (for Berkshire, profitable) insurance operation that provides the float.

Setting aside the value of the active insurance operation and its profits, consider:

A boring new corporate loan is worth nothing, net, since you get the cash and you get the liability. Maybe you'll invest that money well, maybe you'll invest it badly, but the moment you draw the loan it's a wash and adds nothing to the firm's value.

Float by itself is very much like an interest free loan. It's better because of the lack of interest cost. But as with any loan, you still get both the cash and the liability.

Float has a different set of strings attached: you might need to repay quite a big chunk of the principal in any given year, or not: it's not known. You might be able to keep it more or less forever, or it might run off slowly over time. You'll have to keep a lot of it in super liquid paper that on average has zero real return after tax, so some portion of the asset side has zero return for you and is thus worth nothing to you beyond its ability to pay off the offsetting liability.

There is no doubt at all that $100bn in float-generated cash that comes with $100bn in liability, taken together (but without the underwriting operation and its potential profit), is plainly worth less than $100bn in cash with no corresponding deferred liability or payback requirements or liquidity restraints or regulatory restrictions etc. I mean, obviously.

Similarly, the combination is worth much more than zero because some of it can be invested over long periods with a decent return that you get to keep, and you don't have to make interest payments. It's just a matter of assessing how much more than zero, and how much less than an equivalent amount of cash without a corresponding deferred liability and restrictions.

If you think about what each of you is discussing--the float asset, the float asset plus its liability, or both of those plus the underwriting operation--it's likely there isn't much disagreement.

Jim

Rear view mirrors don’t always offer the best view. Especially when the chart highlights an historic equity bull run post GFC.


Our insurance float and cash are sources of optionality. And yeah, a static low return assumption forever is a tempting assumption. Just look at those numbers.

Most of the time they earn—yes modest returns, but their real value lies in the ability to act when extraordinary opportunities appear. A long bull market since 2009 has severely limited these moments, which might make recent returns look representative of the future.

In truth, the optionality embedded in durable, low-no cost float means that when markets dislocate or acquisitions become attractive, we can deploy large amounts of capital at returns well above “normal”.

Over decades, those rare moments dominate compounding, one reason why we maintain liquidity far beyond what day-to-day operations require. Simply looking at average forward returns misses this structural advantage—float is both (for us, free) leverage and flexibility, and that optionality is a central driver of our long-term value.

Optionality has real VALUE.

Our cash, obviously, is worth MORE today than it was 2 days ago. What it buys at the “market store” is more than what it could buy at the market store just 2 days ago. Same with Float. I don’t assume low return float forever.

I don’t assume low return float forever.

Seems reasonable.

But it also seems reasonable to assume that some fraction of it will have no return forever, on average. It's relatively rare for 3-month T-bills to offer a positive return after tax and inflation, and there will always be a lot of bills. Given the scale of the insurance operations these days, maybe $30-40bn floor?? Average over time rather than the floor, probably well over $50bn?

Jim

so let’s just see if I can explain in a way that even YOU can understand

From this it's clear who is offending individual. It's also clear that you are not really interested in a thoughtful analysis of the company. Goodbye!

The two of you had a serious lengthy and productive discussion. Finally valuable posts again here. Then one sentence:
.....you seem to unfortunately resort to unproductive rhetoric.

which might not have been nice, right. But to see that as "condescending insult[s]"? As a German with our reputation in the English speaking world to be so direct it´s bordering on rudeness I´d say: Don´t be so overly sensitive and reactive. No need to pay it back doubly with:
.....let’s just see if I can explain in a way that even YOU can understand

But done, ok, so you two are even and no need for
.....Goodbye

90+% of all posts here are boringly uninteresting. Shoulder-clapping how great Berkshire is, endless always the same wishes for dividends, for buybacks etc. etc. Finally a valuable real discussion you two had. Good!

Pussies. Don´t ever holiday in Germany. You couldn´t stand it.

For those who disagree, you may just be talking at cross purposes: there is a difference between valuing the assets provided by the float, valuing those assets in combination with the offsetting float liability and restrictions, and valuing those two along with the (for Berkshire, profitable) insurance operation that provides the float.



I think you are correct to say that a lot of people disagreeing about the value of the insurance float are not addressing these aspects separately. Buffett himself has not helped by saying things like "the float is better than equity", but his explanation is that the float is not just cost-free, it has negative cost, since Berkshire's underwriting has tended to be profitable.

I think it would be clearer for him to say that the float is valuable, although it is offset by an equal liability, because it is like an interest-free loan that is indefinitely renewed, and on top of that, the insurance business is valuable because not only does it produce investable float, but it also produces underwriting gains, which can be valued separately. But of course, you are not allowed to count those gains twice, first as underwriting profits, and then as negative cost of float; these two are the same thing.

So when Buffett says that he wouldn't accept $1m for a million dollar's worth of float, what he really means is that he wouldn't accept $1m for the insurance business that produces $1m in float. That is because that insurance business is spinning of $30,000 in underwriting gains every year on average (at a CR of 97), PLUS it can be invested, much of it with good returns. Counting these things separately, say as $450,000 worth of underwriting business (at a 15x pre-tax multiple,) and another $700,000 (using Jim's estimate) based on how the float can be invested, is how you get more value than $1 million, but only part of that is from the float.

Post New

US Policy FAQ

---

Contact Shrewd'm
Contact the developer of these message boards.