AdrianC wrote: "Those with time and inclination might be able to tell us mortals what this all means."

CAPE says nothing about Berkshire Hathaway, and nothing about what the market is expected to do in the next year. CAPE can be used to project what the market is expected to do in the next 5 years. This CAPE projection has been wrong recently. Adjusting CAPE for accounting changes that took effect in the early 1990s makes the recent projections more accurate.

Try feeding these prompts into your AI:
A thought experiment: were new investors better off in 1929 or 1999?
What was CAPE in 1929 and 1999?
CAPE was higher in 1999, but new investors did better compared to 1929. Was this just random luck, or does CAPE need an adjustment?

My AI search yielded a couple of interesting ideas on how to adjust CAPE:
>>> NIPA Earnings Substitution: Instead of using manipulated GAAP corporate earnings, analysts substitute National Income and Product Accounts (NIPA) profits from the Bureau of Economic Analysis. NIPA calculations adjust for economic depreciation and remove accounting-driven write-downs.
>>> The Shiller-Black-Scholes Premium: Adjusting the ratio against prevailing risk-free interest rates (which were drastically different in the 1930s versus the post-2000 era) reveals that modern equities often carry a more justified equity risk premium despite higher raw multiples.

My bottom line is I ignore the claims that the market is going to crash because CAPE is at all time highs. CAPE needs adjustment because of historical accounting changes.

=== summary of the paper ===

Accounting distortions changed how earnings were reported. FASB #2 (1974) requires expensing of R&D, and this created a wedge between GAAP earnings and the earnings that investors value. This divergence expanded sharply in the early 1990s, driven by accounting standards changes including SFAS 121 (1995) that required more asset impairment.

"The timing of CAPE's structural break aligns precisely with these accounting distortions becoming economically material... CAPE-H replaces GAAP earnings with adjusted earnings (GAAP earnings minus special items plus R&D)" from the CAPE-H paper:
Palazzo, Dino, The CAPE that Cried Wolf (May 08, 2026). Available at SSRN: SSRN: Research paper

=== links to past posts ===

I obviously agree with the idea that CAPE needs to be adjusted for accounting changes. I came to this conclusion after looking at the CAPE data, and posted messages in December and April:

Author: lizgdal
Subject: Re: Is Morningstar dishonest?
Date: 12/17/25
"Something changed around 1991. Comparing earnings yields before and after 1991 is suspect. In 1991, the S&P 500 price increased 26% while earnings dropped."
lizgdal replies on Is Morningstar dishonest?

Author: lizgdal
Subject: Re: Is Morningstar dishonest?
Date: 12/18/25
"The Shiller CAPE data is more consistent over time with an adjustment: double the earnings starting in 1991."
lizgdal replies on Is Morningstar dishonest?

Author: lizgdal
Subject: Re: Control Panel: Valuations. Temporary?
Date: 04/19/26
"There was a major accounting change around 1991, and so I double the Shiller CAPE values before 1991 when doing historical comparisons. With that adjustment, the recent CAPE value is within the historical range, but is high."
lizgdal replies on Control Panel: Valuations. Temporary?

Author: lizgdal
Subject: Re: Control Panel: Valuations. Temporary?
Date: 04/21/26
"No links, just an observation after looking at the CAPE data. The data shows a break around 1992, and there were various FASB accounting changes around that time. A thought experiment: were new investors better off in 1929 or 1999? CAPE says the 1999 bubble was far worse, but the returns after the bubbles burst were far worse in 1929. Was this just random luck, or does CAPE need an adjustment?"
lizgdal replies on Control Panel: Valuations. Temporary?