Bond ETFs and Price Discovery in Stress

Individual corporate bonds often trade by appointment: a dealer has to be found, a price negotiated, and in a stressed market that process can seize up almost entirely, with bonds going hours or days without a genuine trade. A bond ETF holding hundreds of those same bonds, by contrast, keeps trading continuously on the exchange all day, every day the market is open. In March 2020, this produced something that looked alarming: several large bond ETFs traded at prices several percent below their published net asset value.

The natural read is that the ETF was "broken" or mispriced. The more accurate read is closer to the opposite: the ETF's exchange price was a live, continuously updated estimate of what the underlying bonds were actually worth in a market with almost no other trading happening, while the fund's official NAV — calculated from the last available quotes or matrix-pricing models for bonds that hadn't traded in days — was stale. The ETF was arguably doing genuine price discovery for an asset class where the "real" market had gone quiet, not failing to track it.

This works because the ETF arbitrage mechanism doesn't require every underlying bond to be liquid — it only requires that authorised participants can still assemble baskets to create or redeem, and during the 2020 stress, funds leaned on custom baskets and cash creates precisely to keep that mechanism functioning even when specific bonds were untradeable.

None of this means bond ETF discounts are always benign — a persistent, large discount can also signal that even the ETF's own liquidity is drying up. But the mere existence of a gap between ETF price and stale NAV, during a bond market stress event, is not on its own evidence that the ETF mechanism has failed.

Regulators and researchers who studied the March 2020 episode afterward largely reached that conclusion: the discounts closed quickly once bond trading resumed, and the ETFs that traded at the widest discounts were often the ones whose underlying markets were the most illiquid to begin with, like high-yield corporate bonds, rather than the ETF structure itself being the source of the problem. The episode is now a standard example cited in favor of ETFs as a liquidity tool precisely because the wrapper kept trading and pricing information flowing when the underlying cash bond market briefly couldn't.