Bond fund inflows risk grows as investors rush into fixed income

Bond fund inflows risk is flashing brighter after investors sent a rush of cash into fixed-income mutual funds and ETFs, according to MarketWatch columnist Mark Hulbert, who says the buying spree may be a warning sign rather than a comfort blanket.

Hulbert cited EPFR, a data provider owned by ISI Markets, showing U.S. bond ETFs have recorded net inflows for 59 weeks in a row and in 66 of the past 67 weeks. Across bond mutual funds and ETFs, net inflows over the past year have added up to about 10% of fund assets, he wrote.

That is a sharp contrast with U.S. stock funds and ETFs, where comparable inflows were less than 1% of assets over the same stretch, according to the figures cited by Hulbert.

Why could heavy bond fund inflows be a risk?

Hulbert’s argument is based on a contrarian reading of fund flows: heavy buying can show investors have become too confident, while heavy selling can show pessimism has gone too far. In that framework, crowded enthusiasm for bonds may leave the asset class vulnerable.

Bond funds are pooled investment vehicles that hold debt such as Treasurys, corporate bonds or broader investment-grade portfolios. They can still lose money when bond prices fall, including when yields rise or when investors pull back from a crowded trade.

Hulbert wrote that the recent pace of bond-fund buying has been exceeded in recent years only in 2021, measured as a share of assets. The following year, bonds suffered their worst bear market since 1793, according to a database maintained by Edward McQuarrie, an emeritus professor at Santa Clara University.

He cautioned that one example does not prove a rule, but said longer-term data point in the same direction. Using EPFR data, Hulbert compared trailing 12-month bond-fund flows with the next 12 months of returns for the Bloomberg U.S. Aggregate Bond Index over the past decade and found a meaningful inverse relationship.

Hulbert said the r-squared for trailing bond-fund flows and subsequent bond-market returns was 14.8%, which he described as statistically significant. By comparison, he wrote, the trailing price-to-earnings ratio’s ability to forecast later stock-market returns had an r-squared of 1.9%.

What does the forecast say for bonds and stocks?

Based on the latest flow figures and the historical relationship he analyzed, Hulbert wrote that average investment-grade bonds are implied to lose 0.8% over the next 12 months on a total-return basis. Long-term U.S. Treasurys are implied to lose 3.8%, he wrote.

Hulbert also argued that fear about stock-market valuations has likely pushed some investors toward bond funds instead of equity funds. That shift, in his view, has made stocks look better relative to bonds.

He said he reached a similar conclusion three months earlier using fund-flow data available then. Since that call, through July 27, the U.S. stock market gained 4.0%, while the investment-grade bond market lost 0.9% and long-term Treasurys fell 2.1%, according to Hulbert.

The takeaway from Hulbert’s analysis is not that bond losses are guaranteed. It is that the rush into fixed income has made a supposedly defensive corner of the market look less comfortable than many investors may think.

This story draws on original reporting from MarketWatch.