The Bond ETF Sales Pitch Is Only Half the Story

The bond ETF industry is approaching $3.5 trillion in assets globally, and the fund managers running these exchange-traded funds would like you to know that bonds are back.

Yields are higher than they’ve been in a decade. Liquidity is deep. Active fixed-income ETFs pulled in more than $200 billion last year alone. The pitch, increasingly, is that bond ETFs are no longer a satellite holding to balance equity risk — they are the foundation of a modern portfolio.

There’s a lot of truth to this. The first generation of bond ETFs, tracking broad investment-grade indexes for fees of just 0.03 or 0.04 percentage point, are one of the great deals in finance. They did for retail bond investing what Vanguard Group Inc. did for equities: democratized access to a market that used to require a dealer relationship and a six-figure minimum.

But the part of the bond ETF complex that’s growing fastest isn’t that part. It’s the active and outcome-oriented funds — multisector strategies, flexible income vehicles, securitized credit funds, options-overlay products — that charge 0.30 to 1 percentage point and promise more yield, less duration, or both. And the marketing pitch behind them quietly elides something important. They aren’t really bond funds.

Consider the typical “multisector” active bond ETF, a category that has absorbed a substantial share of recent flows. Its yield looks attractive: well over 5% in many cases, against the four-and-change you get from a broad investment-grade index. Its duration is shorter, too, which sounds like a free lunch in a world where investors fear rising rates. How is the manager doing this?

He is doing it by buying things that aren’t really bonds in the way an investor sitting at a kitchen table thinks about bonds. The yield uplift comes from below-investment-grade credit, collateralized loan obligations, non-agency mortgages and emerging market debt. These are securities whose prices fall hard when equities fall hard — exactly when an investor most wants the bond sleeve to hold its value. They are, in useful shorthand, equity-light. You are paying a manager to take equity-shaped risk and put it in a wrapper labeled “bond fund.”

This isn’t a scandal. It’s a perfectly defensible portfolio construction, and the prospectuses say what they say. The problem is that the products are sold against a benchmark — the broad index — that is not equity-light, and against a category — fixed income — that has connoted safety in retail investors’ minds for 40 years.

A standard mean-variance optimizer — the textbook tool for choosing portfolio weights — fed the historical return and volatility of one of these funds will report low correlation with equities and recommend a hefty allocation. The optimizer is fooled by the wrapper. In stress, when correlations all run to one, the equity-light part of the bond sleeve is going to behave like equities, and the investor who bought it for ballast is going to be unpleasantly surprised, like the guy who gets a notice that his homeowner’s policy is cancelled while the hurricane is approaching.