“No other fund manager made more money for people than Bill Gross,” wrote Karen Dolan of Morningstar Inc. when naming the bond king the fixed-income manager of the decade in 2010, and she could have extended that to a 35-year title. Now Gross tells us that if we hear people bullish on bonds, “they just wanna sell you a bond fund,” and that the “total return” philosophy he originated at Pacific Investment Management Co. in the depths of 1980s’ bear market is dead.

Total return is based on Bond Math 101. If a 10-year bond is selling at a 0% yield it has a duration of 10, meaning a 1% increase in interest rates will cause it to lose about 10 x 1% or 10% in price (the exact loss is 9.47%). Back in the early 1980s, the 10-year Treasury note was available at a yield of 15.84%. That gave it a duration of five years, which meant rates could go up to 19.50% and the interest on your bond would cover the loss in price. Absent a massive increase in rates to unprecedented levels, you had to make money.

The higher the yield at which you purchase the bond, the lower the duration so the less sensitive you are to interest rate increases, and the more income you have to offset them. Moreover, the 10-year Treasury is just a benchmark for bond investing, Gross was famous for squeezing out extra yield using corporate bonds, mortgages, derivatives and other instruments.

But Gross points out that’s not the whole story. When yields are high it likely means expected inflation is high, or yields are expected to increase, or both. The chart above shows the actual 10-year annualized inflation-adjusted return that investors in 10-year Treasuries earned versus the yield at the time of bond purchase. Gross’ 38 years at Pimco are shown in yellow, all other years are in blue.

The yellow dots illustrate the total return argument. After inflation, investors in 10-year Treasuries earned on average about 75% of the yield at the time of their purchase minus 1%, usually within a percent of the predicted amount. The exception was some negative returns near the end of Gross’ Pimco tenure. They occurred at very low interest rates and began in the Spring of 2012. By the Fall of 2014, Gross and Pimco parted ways. Of course, investors in Gross’ Total Return Fund did considerably better than investors who bought 10-year Treasuries, but no bond manager can buck the market, only add a moderate premium to whatever the Treasury gives.

Note that the total return story becomes undependable outside Gross’ Pimco years. Realized real returns could be very high or negative, with no obvious relation to purchase yield. Gross is telling us that we’re in that state now, and the fundamentals make a low or negative return likely.

The 10-year yield was about 4.59% at the time of writing, duration was eight years, and rates could rise 59 basis points to 5.18% to wipe out one year’s income. The 10-year breakeven inflation rate was 2.37% and 10-year Treasury note futures suggest the market expects stable yields for the near term. Based on this, it might seem like the total return approach remains sound. The yield is a comfortable premium over inflation, and bond investors can boost that further by using instruments other than Treasuries. In similar scenarios during Gross’ Pimco years, investors made between 2.5% and 4% above inflation from 10-year Treasuries.

So why does Gross think total return is dead? It’s not new, it’s a theory he came up with in 2013 when the strategy was failing for the first time since the late 1970s. As he explained, if total credit does not expand enough to pay the interest on existing credit, capital is subtracted from the economy. Debtors cannot borrow enough new money to pay interest, so they must pay it out of income. With $77 trillion in total U.S. credit at an average interest rate of 5.5%, that’s $4.2 trillion additional credit needed.

In a healthy economy, business expansion and household optimism could demand its share of that $4.2 trillion. But in recent times business and household debt has been growing at only 1% to 2%. The government has made up the difference, expanding Treasury debt by 10% per year to pay interest on existing borrowings and fund deficits. According to Gross, it must continue doing this to keep the economy growing at a nominal 5.5% rate.

Unfortunately, an increasing supply of Treasuries will push down prices—which means yields go up—and could fuel inflation. Rising yields and inflation kill the total return argument.

While Gross’ analysis makes sense, I see two alternatives. If things go along more or less as expected, current 10-year Treasury yields should produce pretty good 10-year inflation-adjusted returns, and high-yield fixed-income investments should do even better. While Gross makes a good case that the monetary situation is unstable, things often seem to chug along as expected for very long periods. Moreover, when things do get bad investors often flock to Treasuries—although not other fixed-income assets—as a safe haven. I don’t see any spark on the horizon to ignite a bond disaster, although I agree with Gross there’s a lot of deadwood to burn if things do go wrong.

The other bull case is that the real economy expands, and creditworthy businesses demand more credit even at higher rates; the government tames deficits; investors retain enthusiasm for bonds; and consumer confidence—and creditworthy borrowing—increase. A lot must go right for that scenario—it has equity-like risk—but it could result in fixed-income providing equity-like returns. In both the 1920s and the 1980s, fixed-income markets seemed even more precarious than today, and Treasuries delivered excellent inflation-adjusted returns for investors like Gross willing to bank on total return. And Bill Gross’ 2013 decision to abandon total return and short bonds is usually suggested as the reason for his non-amicable departure from Pimco.

I don’t wanna sell you a bond fund, but I’m not going to tell you to sell your bond funds either. Bill Gross knows more about bonds than anyone and investors should pay attention. But to me, diversification is the emperor that outranks kings. Getting out of bonds means taking risks somewhere else, and most investors are wildly overexposed to equities. It seems sensible to me to reallocate some fixed-income to non-U.S.-dollar bonds, or shorter duration, or otherwise reduce exposure to U.S. interest rates. Total return may be dead, but bonds aren’t.

Aaron Brown is a former head of financial market research at AQR Capital Management. He is also an active crypto investor, and has venture capital investments and advisory ties with crypto firms.