Stocks Keep Rising While Getting Cheaper
The S&P 500 closed above 5,900 on May 23, 2026, marking its sixth consecutive week of gains. Yet forward price-to-earnings ratios have actually compressed over the past twelve months. The paradox, stocks climbing in pric
The S&P 500 closed above 5,900 on May 23, 2026, marking its sixth consecutive week of gains. Yet forward price-to-earnings ratios have actually compressed over the past twelve months. The paradox, stocks climbing in price while becoming more affordable relative to earnings, tells a story about corporate profitability that most financial commentary ignores. It also raises a question Austrian economists have asked for over a century: are these gains real, or are they an artifact of monetary distortion?
The Paradox of Cheaper Expensive Stocks
Anthony Pompliano's latest letter laid out a straightforward observation. The S&P 500 has risen roughly 15% over the trailing twelve months, yet the index's forward P/E ratio has dropped from approximately 22x to around 20x. That compression means earnings expectations grew faster than prices. Companies are producing more profit per share than the market anticipated, and the stock prices have not fully caught up.
This is not a normal pattern. In most bull markets, rising prices pull valuations higher. Investors pay more for each dollar of future earnings because momentum breeds optimism. The current environment flips that script. Earnings revisions have turned sharply upward, driven primarily by the technology sector, energy margins, and financial services. S&P 500 earnings per share are now projected near $275 for 2026, up from roughly $240 a year ago. The denominator in the P/E equation is growing faster than the numerator.
For traditional equity analysts, this is unambiguously bullish. A market that grows cheaper as it rises has room to run. The historical average forward P/E for the S&P 500 sits near 17x. At 20x, valuations remain elevated but not extreme. Compare that to the dot-com peak of 25x or the post-pandemic spike above 23x. The math suggests equities could absorb another 10-15% in price gains before hitting the kind of valuation levels that historically trigger mean reversion.
Earnings Growth Under the Hood
The earnings story is not evenly distributed. Strip out the top ten companies by market capitalization, the so-called Magnificent Seven plus a rotating cast of AI-adjacent firms, and the remaining 490 stocks in the index show much more modest earnings growth, roughly 6-8% year over year. The concentration of profit growth in a handful of mega-cap technology companies continues to warp the aggregate picture.
Nvidia alone is expected to generate over $130 billion in revenue for its fiscal year ending January 2027, a figure that was barely conceivable three years ago. Apple, Microsoft, and Alphabet continue to print cash at rates that dwarf most national GDPs. Meta's advertising machine delivered $165 billion in trailing revenue as of Q1 2026. Amazon Web Services crossed $110 billion in annualized revenue.
These are real businesses producing real cash flows. The bears who have called this a bubble for three years have been wrong on timing and arguably wrong on substance. AI infrastructure spending is translating into measurable productivity gains across enterprise software, cloud computing, and digital advertising. Whether this spending continues at its current pace is a legitimate debate. Whether it has produced tangible earnings growth is not.
The skeptical case rests on sustainability. Capital expenditure by the five largest technology firms exceeded $250 billion in 2025. That spending must eventually produce returns sufficient to justify the investment. If AI spending plateaus or if the productivity gains fail to compound, the earnings trajectory flattens, and those compressed P/E ratios snap back upward. The market is pricing in continued acceleration. History suggests acceleration rarely lasts as long as investors hope.
The Monetary Backdrop
No honest discussion of equity valuations can ignore the Federal Reserve. The federal funds rate sits at 4.25-4.50% as of May 2026, down from the cycle peak of 5.25-5.50% in 2023. The Fed has delivered three 25-basis-point cuts since September 2025, with markets pricing in two more by year end. The balance sheet, while smaller than its $9 trillion peak, still holds over $7 trillion in assets. Quantitative tightening has slowed to a crawl.
Cheap credit inflates asset prices. This is not a controversial statement. It is a mechanism that the Fed itself acknowledges through the "wealth effect" channel of monetary policy. When the central bank suppresses interest rates below what a free market would produce, capital flows into risk assets. Stock prices rise not because companies are worth more in any absolute sense, but because the alternative, holding cash or bonds, pays less.
Ludwig von Mises called this the "crack-up boom," a phase where asset prices surge because market participants realize that the monetary authority will continue to expand the money supply. Friedrich Hayek's business cycle theory predicted that artificially low interest rates would distort the capital structure, directing investment toward longer-term, capital-intensive projects that appear profitable only because of the distorted price signal.
The current environment fits this framework uncomfortably well. Corporate earnings are growing, yes. But they are growing in nominal dollar terms, measured against a currency that has lost roughly 25% of its purchasing power since 2020 according to the Bureau of Labor Statistics' own CPI data. A stock market that rises 15% in nominal terms while the dollar loses 4-5% in purchasing power annually is delivering real returns that are far more modest than the headline numbers suggest.
Bitcoin as the Honest Denominator
This is where the story connects to sound money. Priced in Bitcoin, the S&P 500 tells a very different story. Bitcoin traded near $109,000 in late May 2026, having appreciated roughly 50% over the trailing twelve months. The S&P 500, priced in BTC, has actually declined. Stocks are not "getting cheaper" in Bitcoin terms. They are getting more expensive while delivering less real value.
This framing is not an academic exercise. It reveals the fundamental dishonesty of measuring wealth in a unit that the issuer can produce at zero marginal cost. The Federal Reserve can create dollars with a keystroke. No one can create Bitcoin beyond the protocol's fixed schedule. When you measure corporate earnings in a unit of account that is being deliberately debased, "earnings growth" becomes a partially illusory phenomenon, a mixture of genuine productivity gains and monetary inflation.
Bitcoin's 21 million supply cap makes it the only widely traded asset with a perfectly inelastic supply curve. Gold comes close, but annual mine production of roughly 3,500 tonnes adds about 1.5% to above-ground supply each year. Bitcoin's current annual issuance rate is approximately 0.85% and will continue to decline with each halving event. By 2028, it drops below 0.4%.
For investors trying to preserve purchasing power across decades, the denominator matters as much as the numerator. A stock portfolio that compounds at 10% nominally while the currency depreciates at 5% delivers 5% in real terms. A Bitcoin allocation that appreciates at 30-50% annually, as it has averaged over any four-year cycle in its history, offers a fundamentally different proposition. The risk profile is different, obviously. The volatility is real. But the direction of travel has been consistent for fifteen years, and the monetary properties that drive it are getting stronger, not weaker.
The Valuation Trap
Pompliano's core argument, that stocks remain attractively valued because earnings are outpacing prices, deserves scrutiny from another angle. Forward P/E ratios depend on analyst estimates, and analyst estimates are notoriously biased toward optimism in bull markets. The consensus estimate for S&P 500 earnings has been revised upward for six consecutive quarters. That streak is unusual. It also tends to precede periods of disappointment.
The 2007 analog is instructive. In mid-2007, the S&P 500's forward P/E sat near 15x, well below historical averages. Analysts projected robust earnings growth for 2008 and 2009. The market looked cheap by every traditional metric. Within eighteen months, actual earnings had fallen 40%, the "cheap" P/E expanded to over 70x, and the index had lost half its value. The lesson: forward earnings estimates are a trailing indicator disguised as a leading one. They reflect the current business cycle, not the next one.
This is not a prediction that 2026 will repeat 2007. The balance sheets of major banks are stronger. Consumer debt ratios are lower. The AI productivity tailwind is real in a way that the housing-driven earnings of 2006-2007 were not. But the principle holds. Cheap-looking markets can become expensive very quickly when earnings miss expectations. And earnings miss expectations most often at precisely the moment when everyone agrees the outlook is favorable.
The contrarian position is not that stocks will crash. It is that the margin of safety is thinner than the P/E compression suggests. A 20x forward multiple is not cheap in absolute terms. It is only cheap relative to the 22x level of a year ago. Relative cheapness is a weaker foundation than absolute cheapness, especially when measured against an interest rate environment that remains restrictive by the standards of the post-2008 era.
What to Watch
Three indicators will determine whether this "cheaper while rising" trend continues or reverses.
First, Q2 2026 earnings reports, beginning in mid-July. If the mega-cap technology firms deliver another quarter of 20%+ earnings growth, the P/E compression story holds. If AI-related spending shows any sign of deceleration, the entire forward estimate curve shifts downward, and the "cheap" market suddenly looks expensive.
Second, the Fed's June and September meetings. Markets are pricing two more rate cuts by December. If inflation data reaccelerates, those cuts evaporate, and the discount rate applied to future earnings rises. A 50-basis-point upward shift in the ten-year Treasury yield would, all else equal, add roughly 1.5 turns to the fair P/E multiple, pushing the index from "reasonably valued" to "stretched."
Third, Bitcoin's behavior relative to equities. Over the past two years, Bitcoin has increasingly served as a leading indicator for risk appetite and monetary conditions. When BTC breaks to new highs, equities tend to follow within weeks. When BTC stalls or corrects, equity markets often weaken. The correlation is imperfect but strengthening. Bitcoin at $109,000 suggests the liquidity environment remains supportive. A sustained move above $120,000 would likely coincide with the next leg higher in equities. A drop below $95,000 would flash a warning signal that the monetary tide is turning.
The stock market may indeed be getting cheaper as it goes higher. But "cheaper" is a relative term that depends entirely on which ruler you use to measure it. If your ruler is the U.S. dollar, the case is persuasive. If your ruler is Bitcoin, the picture looks very different. Choose your denominator carefully.
Source: Pomp Letter