On January 16, this writer held a Zoom call with Rob Arnott to get the equity market savant's take on a reliable metric that's flashing red for U.S. big cap stocks. For this writer, Arnott's the best go-to source for assessing when and where equities are way under or overvalued. As founder and chairman of Research Affiliates, a firm that oversees investment strategies for $156 billion in RAFI mutual funds and ETFs, Arnott's proven a winner in the real-world contest of developing strategies that deliver benchmark-beating returns for big institutions and retail customers alike. Plus, he's assembled that record by harvesting ideas from the best minds in academia, from his time as editor-in-chief of the Financial Analysts Journal to today's brainstorming sessions enlivened by RA's long roster of PhDs.
A prime example of turning a concept born of research into a durable, practical blueprint for enriching investors: His role as father of "fundamental indexing," an approach that allocates the shares in a RAFI fund not by market cap, but by the size of the companies comprising the index in the overall economy. RA assigns those weights based by such metrics as revenues, net worth and dividends—a formula that drives dollars away from what's hot and expensive towards the cheap and unloved.
Rob called in from Las Vegas, where he just welcomed the arrival of his second grandchild—and the burly, bearded, always exuberant money manager was in a celebratory mood.
But when I asked Arnott whether I was right in thinking that a favorite measure among economists for tagging times when stocks are radically cheap or too richly valued called the Equity Risk Premium was pointing to deep danger, Rob's view was anything but cheery. "The extra return stocks offer versus bonds has shrunk to the point where it's extremely low by historical standards," he explained. "That's what the ERP shows, and it's a key guide to future returns. At these levels, the ERP suggests that we're in another bubble, not one where stocks are as recklessly priced as during the 2000 frenzy, but a high-risk scenario where they're pretty aggressively priced." Though he believes that a steep fall isn't necessarily imminent, the ERP's big shift in the wrong direction greatly increases the odds a bear market will strike relatively soon—despite the overwhelmingly bullish sentiment reigning on Wall Street that's only intensified now that Trump's victory promises a new pro-profits era of corporate and personal tax cuts and sweeping deregulation.
What is the Equity Risk Premium?
In basic terms, the ERP is the difference between the long-term, inflation-adjusted returns the market expects for stocks versus the expected returns for the risk-free alternative, government bonds, again, minus anticipated inflation. A good figure for the projected "real" gains on equities is the "earnings yield," the dollars in GAAP net earnings that the S&P 500 delivers for every $100 investors pay for their shares. (That's also the inverse of the price-earnings ratio; the higher the PE goes, the fewer Corn Flakes you're getting in the box.) The earnings yield takes account of rises in inflation because corporate profits are mainly CPI-protected: Companies regularly lift their prices to offset the increases in wages and materials costs—it's the rise in what they charge for groceries, rent and gasoline that constitute inflation.
The expected "real" return on bonds equals the yield on 10-year Treasury Inflation-Protected Securities or TIPs. That number represents the extra percentage points, over and above projected increases in the CPI, that 10-year Treasuries, at their current rates, are promising over the next decade. So the ERP is simply the earnings yield, what stock are promising after inflation, minus the yield on 10-year TIPs, what long-term Treasuries are offering over inflation.
Comparing the "real" numbers is so essential because they represent how much investors can anticipate their purchasing power will grow in the years ahead. If the dollars you're getting simply match the overall increase in prices over a decade, you can't buy any more goods and services with that money ten years from now than you do today. The bigger the premium over forecast inflation a 10-year Treasury is paying, the more appliances, restaurant meals, and cars you can purchase with the coupons—that's the "real" deal, the spread over the CPI that grows your annual spending power year after year. And government bonds are offering that inflation-beating edge with no risk of losing your capital, if you hold them to maturity.
By contrast, equities are highly volatile, so they only become much more attractive than Treasuries when their inflation-adjusted future returns far exceed the edge that the 10-year, say, is offering over and above the projected increase in the CPI.
As Warren Buffett points out, stocks and especially government bonds are locked in constant competition for investors' money. And stocks are losing the contest. Today, the ERP—the extra margin offered by stocks over Treasuries—has reached its lowest level since the tech bubble of 2000. Historically, a narrow ERP's been an extremely negative signal for stocks. But Wall Street isn't paying attention. Overall, analysts and strategists predict that the S&P 500 will reach 6,666 at the end of 2025 for an annual gain of 12.5%. You'll seldom hear the ERP mentioned by guests from the banks and brokerages on the TV business shows, maybe because it contradicts their upbeat scenario. Or perhaps because it sounds like a wonky concept advanced by academics that seems irrelevant as a forecasting tool since even as the ERP gets worse and worse, stocks keep soaring.
But as Arnott explains, "It just makes sense. When Treasuries that provide guaranteed income are offering you almost the same return as stocks that are extremely pricey and have loads of risk, people will eventually dump stocks until their prices fall to the point where they regain their edge." That adjustment often happens after a lag, but the basic market math dictates that it's inevitable. So let's examine how a big ERP opened the gates for the giant rally in recent years, and how the great shrink foreshadows a dim future not for all big cap U.S. stocks, but the S&P overall, and especially the glamor names in tech that led the immense rally.
The ERP's suffered a big squeeze as rates spiked while the S&P kept soaring
In the aftermath of the Great Financial Crisis and beyond, the environment proved highly favorable for the Equity Risk Premium. The main reason: The super-low real rates engineered initially by Fed Chairman Ben Bernanke to rescue America from the Global Financial Crisis followed by successive new floods of liquidity designed by his successor Jerome Powell to combat the Trump tariffs in 2019, then counter the downdraft from the COVID crisis the following year. From 2012 to early 2022, the inflation-adjusted rate on the 10-year Treasury averaged a puny 0.25%, compared to around 3% in the 1990s and around 2.5% from 2004 to 2008. Over the decade ending in February of 2022, the PE also cooperated for the most part. The S&P multiple rose gradually, but averaged roughly 20, for an earnings yield of 5%. Result: The expected return on stocks boasted a gigantic margin over the ten-year of 4.75% (the 5% earnings yield minus the 0.25% real rate). The biggest factor in creating that fat ERP: those astonishingly low rates.
The sumptuous ERP created a strong tailwind for stocks, and the S&P 500 climbed 250% over that golden decade. The signs of overheating got serious in early 2022 when two forces converged to start compressing the ERP. First, S&P prices kept rising faster than profits, swelling the PE and pounding the earnings yield. Second, real rates went on one of the fastest, steepest ramps in history. In early February of 2022, the Fed's expansionary policies at their apex, the inflation-adjusted yield on the 10-year sat at a minus 0.61%. As of January 17, it's vaulted to 2.17%.
Of course, the headline or "nominal" 10-year yield you see that flashes on your TV screen and gets all the attention is just the sum of expected inflation plus the real rate. In that nearly three year timeframe, the headline rate jumped from 1.8% to 4.58%. But as of today, the Fed's brought expected inflation down to around the same modest level as before the outbreak that started in early 2022. As a result, the jump in the 10-year Treasury yield isn't coming because the market thinks future inflation will be worse today than investors believed in early 2022, due to the Fed's restrictive policies. What's driven the entire increase from 1.8% to 4.58% is a jump in the real rate of 278 basis points or almost 3%. That's the biggest single story in the recent history of equity markets. And it's scary.
As the real rate spiked, the earnings yield kept falling. As of mid-day on January 21, the S&P 500 was hovering near an all-time record at 6,035. Based on GAAP trailing 12 month earnings of $200.27, the PE towered at 30.13. That's its highest reading at any period in the past quarter century, except when earnings collapsed during the GFC and pandemic. The PE of just over 30 puts the earnings yield at 3.33%. So the current ERP, the lead of stocks over bonds, is a mere 1.16% (the 3.33% earnings yield less the 2.17% real rate on the 10-year). For every $100 they invest, folks and funds can now either collect $2.17 a year on TIPs risk free plus inflation or $3.33 on dicey big caps plus inflation. As Arnott puts it, "Why would they choose stocks posing all that risk for just an extra dollar?"
Arnott & Co. are positing that a fall in the multiple from today's super-high of 30 will lower the S&P's real return over the next decade to around 0.9% per year, well below the 2.17% comparable rate on the 10-year. That puts the ERP in negative territory. In other words, the combination of huge valuations and the most alluring sustained inflation-adjusted yield in many years is exercising a classic squeeze that's driven the the advantage of extremely unpredictable stocks over ultra-safe bonds to less than nothing. By Arnott's math, long-term Treasuries will give investors almost 1.3 points more in annual purchasing power in the years ahead than the S&P 500.
Arnott stresses that it's the excessive sway of the Mag Seven and other highly expensive tech stocks that's mainly responsible for the index's paltry ERP. "Over thirty-five percent of the S&P's valuation rests in just the seven top names," he says. "That's the highest concentration since the 19th century when a few giant railroads and oil companies dominated the market. Those tech giants account for only 18% or 19% of the U.S. economy but are nearly 40% of the S&P's market cap. Their valuation is over twice the size of their economic footprint." You can see why Rob loves fundamental indexing that avoids empowering high-flyers to wag the index.
Arnott sees big opportunities in such areas as emerging markets and European equities that are offering bargain PEs and as a result of higher ERPs that provide plenty of space for prices to advance. And he thinks that sundry U.S value stocks remain reasonably priced. But he believes that the S&P 500 as a whole features just about the world's least promising basket of equities right now. Arnott notes that in most decade-long economic cycles, bear markets hit every five to seven years. "So there's about a 20% chance you'll get a sharp downturn in any one year," he says. "But now, I'd put the chances much higher for both 2025 and 2026, at about 50% each. And the reason is the extremely low or maybe below zero equity risk premium."
It's a message Wall Street doesn't want to hear, and President Trump—who often measures his policies' economic success by the stock market's performance—won't want to hear, either.
