There is a lot of bear noise right now, and noise is the right word for most of it. But two of the indicators behind that noise are not noise — they are real, and dismissing them because the people shouting about them are often wrong about timing would be its own kind of mistake.
Most bear cases right now are not worth your time. Vague warnings about "unsustainable valuations," recycled dot-com comparisons with no specific mechanism attached, and permabears who have called eleven of the last two corrections — that is noise, and NGE's job is not to amplify it. But underneath the noise, two specific, measurable numbers are genuinely unusual by historical standards, and an honest letter has to say so plainly rather than dismiss everything bearish as equally unserious.
The S&P 500's Shiller CAPE ratio — price relative to ten years of inflation-adjusted earnings, designed specifically to smooth out short-term noise — is near 40. In the full history of the index, that level has been reached exactly twice: the dot-com peak, where it hit roughly 44 before the index lost most of its value over the following two years, and now. The long-run average sits around 17. A ratio near 40 is not a forecast of what happens next. It is a statement that investors are currently paying a historically rare price for a dollar of smoothed earnings, and that fact does not stop being true just because the market keeps making new highs.
The second number is the Buffett Indicator — total US market capitalization relative to GDP — currently near 219%. Warren Buffett himself set the reference point for how seriously to take this: when the ratio approached 200% in 1999 and again in 2000, he said plainly that investors "are playing with fire." The market is now further into that territory than it was at the dot-com peak he was describing. Buffett's own recent behavior is consistent with that discomfort without being a clean bearish signal — Berkshire Hathaway has been accumulating record cash alongside $4.5 billion in share buybacks, which is a genuinely mixed message, not a unified one: buying back your own stock while sitting on unusually large cash reserves is not the position of someone who thinks everything is cheap, but it is also not the position of someone in full retreat.
None of this makes the bear case airtight, and the strongest bull arguments right now are genuinely strong, not just louder. 2026 corporate earnings are tracking to grow roughly 30% over last year — an extraordinary real number, not a hoped-for one. Nominal GDP growth is running above 6%. Roughly $750 billion is flowing into the economy through AI-related capital expenditure, much of it debt-financed but landing as real activity rather than speculative froth alone. Corporate bond spreads remain benign, which matters more than almost any other single credit-market signal: spreads widen before crises far more reliably than equity valuations predict them. Ned Davis Research's own Secular Bear Watch model — built specifically to flag the start of extended bear markets — has not triggered. Its own chief global strategist's conclusion, as of last week: the report does not indicate that a secular bear has started, though the market is "overbought, overowned and overvalued," conditions that have historically preceded prior secular tops without being sufficient on their own to cause one.
The dot-com comparison is the one bears reach for automatically, and it is also the one that most needs a specific caveat rather than a blanket endorsement. DoubleLine's Jeffrey Gundlach made the sharpest version of this point directly: on traditional valuation measures, today's market is about as overvalued as it was at the last bear market's starting point — but bond yields now sit roughly 400 to 500 basis points higher across the curve than they did then. That is not a small footnote. A given equity risk premium means something different when the risk-free alternative pays meaningfully more than it used to. Comparing today's CAPE ratio to 1999's without adjusting for the interest-rate backdrop both were priced against is comparing two numbers that are not actually measuring the same thing.
A rare valuation signal is not the same claim as an imminent crash, and treating the two as interchangeable is how both sides end up talking past each other. The CAPE ratio and the Buffett Indicator are telling you something true about price today. They are not telling you when, or whether, that price gets corrected — and nobody credible has ever claimed they could.
In keeping with this letterhead's own standard — a claim has to be specific enough to be provable wrong — here is what would move this from "valuations are historically stretched, proceed accordingly" to genuine alarm: corporate bond spreads widening meaningfully from current benign levels, the $750 billion AI capex figure showing signs of being financed by deteriorating credit quality rather than healthy balance sheets, or 2026 earnings growth coming in well below the 30% currently tracked. None of those three has happened yet. All three are worth watching specifically, not generally.
The bears making vague, mechanism-free warnings are noise, and should be treated as such. The two specific numbers underneath the noise — a CAPE ratio near its second-highest reading ever, and a Buffett Indicator above the level its own namesake flagged as dangerous — are not noise, and pretending otherwise because the market keeps grinding higher is its own form of denial. The honest position is neither "ignore the bears" nor "the crash is coming." It is: valuations are genuinely rare by history's own measure, the macro backdrop underneath them is genuinely healthier than 1999's was, and the specific evidence that would resolve this tension — credit spreads, capex financing quality, earnings delivery — hasn't spoken yet. Watch those three, not the noise.