ICI Viewpoints

ICI Viewpoints
Share

ETFs Don’t Move the Market—Information Does

By Shelly Antoniewicz

There they go again.

After last week’s turmoil in the bond market, some commentators sought to link net redemptions from bond ETFs to heightened volatility in the U.S. Treasury market on Tuesday, March 4. But as we have written before when examining the relationship between equity ETFs and stock market volatility, such assertions break down under a careful examination of the data, because they downplay the primary driver of volatility—news that affects investors’ investment decisions.

A Small Slice of the Pie

The net issuance of Treasury bond ETFs represents a tiny share of trading in the $12 trillion U.S. Treasury market, which is widely considered one of the deepest, most liquid markets in the world. Hundreds of billions of dollars worth of Treasury securities change hands every day, so it is highly unlikely that the $6.6 billion in net redemptions of Treasury bond ETFs that occurred on March 4 applied much pressure to this market.

Here’s what happened. On March 4, net redemptions of bond ETFs were concentrated in bond ETFs that primarily invest in Treasury securities with remaining maturities in two ranges—between one and three years, and between three and seven years. The bond ETFs identified by ICI Research that fit one of these two investment descriptions collectively had $6.7 billion in net redemptions on March 4: $3.4 billion in one- to three-year Treasury bond ETFs, and $3.3 billion in three- to seven-year Treasury bond ETFs.

Net redemptions in all Treasury bond ETFs were $6.6 billion on March 4, meaning that there were actually net inflows into some Treasury bond ETFs.

According to data published by the New York Federal Reserve, the average transaction volume in Treasury securities with remaining maturities between one and seven years has been $326 billion per day since the start of 2014. As a result, if the authorized participants of these Treasury bond ETFs sold the securities they received from their redemptions on March 4, their sales would have accounted for just 2 percent of the average daily transaction volume in that part of the Treasury yield curve.

A Tenuous to Nonexistent Link

News reports on the impact of bond ETFs focused on the 8.4 basis-point increase in the yield on the 10-year Treasury bond on March 4. But movement in the 10-year Treasury bond yield is hardly the relevant measure when—as in this case—net redemption activity all occurred in the one- to seven-year maturity range. Instead, we should examine yield changes in Treasury securities in this range.

On March 4, the yield on the Barclays Capital U.S. 1–3 Year Treasury Bond Index increased by 2.5 basis points, while the yield on the Barclays Capital U.S. 3–7 Year Treasury Bond Index rose 6.4 basis points—both smaller than the increase in the 10-year Treasury bond yield on that same day. In other words, the rate increases (driven by price declines) were larger in longer maturity Treasuries (i.e., 10 years), where there was less ETF activity.

In addition, other data suggest that the link between net redemptions of these Treasury bond ETFs and price impacts on their matched-maturity Treasury securities is tenuous to nonexistent. As shown in the table below, there have been days within the past couple of months in which the yield on the Barclays Capital U.S. 1–3 Year Treasury Bond Index has increased nearly as much, if not more—with little to no net redemptions in Treasury bond ETFs in these maturities.

For example, as recently as February 28, the yield on the Barclays Capital U.S. 1–3 Year Treasury Bond Index increased by 2 basis points and the yield on the Barclays Capital 3–7 Year Treasury Bond Index increased by 5 basis points. However, only the one- to three- year Treasury ETFs had net redemptions and the magnitude of those redemptions was small—only $67.6 million. In contrast, the three- to seven-year Treasury bond ETFs had $12.2 million in net share creations and certainly could not have contributed to the rise in yields on three- to seven-year Treasuries that day.

In another instance, on January 8, the yield on the Barclays Capital U.S. 1–3 Year Treasury Bond Index increased by 4.1 basis points and the yield on the Barclays Capital 3–7 Year Treasury Bond Index increased by 8.2 basis points, but there were no net redemptions in corresponding Treasury bond ETFs on that day. In fact, $12.8 million in net new shares of one- to three-year Treasury bond ETFs were created on that day. Notwithstanding the relatively small dollar amount, these share creations would have served as source of demand for the underlying Treasury securities, and acted as a counterbalance to the downward price pressure on one- to three- year Treasuries.

Examples of Comparable Yield Changes in the Treasury Market

 

Date

One- to three-year Treasury maturity

Three- to seven-year Treasury maturity

One-day yield change1
Basis points

Net share issuance2
$ millions

One-day yield change3
Basis points

Net share issuance2
$ millions

1/8/2014

4.1

$12.8

8.2

0

1/14/2014

2.4

$2.5

5.3

0

2/11/2014

2.0

-$7.2

4.4

0

2/28/2014

2.0

-$67.6

5.0

$12.2

3/4/2014

2.5

-$3,448.8

6.4

-$3,266.7

1 Yield of Barclays Capital 1–3 Year U.S. Treasury Bond Index.
2 Dollar value of gross issuance minus gross redemptions of shares. A positive number reflects net creations of ETF shares. A negative number reflects net redemptions of ETF shares.
3 Yield of Barclays Capital 3–7 Year U.S. Treasury Bond Index.
Sources: Investment Company Institute, Bloomberg, BlackRock

ETFs Are Just One Way to Trade

Even if Treasury bond ETFs didn’t exist, investors could easily influence yields on Treasury securities by other means, such as purchasing Treasuries through brokers and dealers or taking long or short positions in Treasury futures. Investors who want out of Treasuries can still sell, leading to exactly the same outcome on Treasury yields, with or without ETFs.

ETFs Are a Means to an End

Most importantly, a change in investors’ views (in this case, prompted by political events in Ukraine) caused the Treasury market to move, not Treasury bond ETFs themselves.

It is important to distinguish between the means by which investors take action in response to news (e.g., through ETFs) with the cause of movement in the market (e.g., news about the political situation in Ukraine). Changes in investors’ views cause market moves. Market prices can be expected to change when relevant new information comes to light. The more surprising the information, the larger the price change, and the more efficient the market, the faster the information will be reflected in the price. This is a well-known feature of financial markets.

In this case, investors were responding to an easing in tensions over the political situation in Ukraine by unwinding exposure to U.S. Treasuries that had been previously acquired in a flight-to-safety. That put downward pressure on Treasury bond prices across the maturity spectrum and boosted Treasury yields. New information moved the market—not Treasury bond ETFs.

Shelly Antoniewicz is senior director of financial and industry research at ICI.