Commentary: ETFs are the new stocks — mind the creation/redemption gap

Exchange-traded funds are synonymous with passive investing. A basic passive portfolio only requires a handful of broad funds to cover most asset classes and countries. Yet, there are now more than 2,500 U.S.-listed ETFs. Most ETFs are closet active vehicles targeting a sector, theme, country or strategy. For example, one of the fastest-growing funds in 2021 has been the Financial Select Sector SPDR Fund (ticker: XLF). It has received approximately $9.5 billion in inflows, expanding its units outstanding by 35% year to date (Oct. 31).

XLF is passively managed, diversified and low cost. But buying XLF or any sector fund is not a passive choice because a broad index investor already has proportionate exposure to financial companies. It is an active decision that financial stocks will outperform the market. Trading XLF replaces stock picking with sector picking. This is also true for thematic and country funds — they are traded to express active views. Narrow products run contrary to the "own everything" approach that the pioneers of passive championed.

We estimate investor activity and the result of their market timing by studying money flows in and out of ETFs. Our analysis indicates that ETF investors have similar behavioral biases of return-chasing that mutual fund data has long showed. An ETF structure might be passive, but humans are not.

When you purchase an ETF, the seller can be an existing holder. However, net cash inflows typically result in new units being created by market makers. Conversely, net cash outflows usually result in units being redeemed. The number of units/shares outstanding is reported daily. A reasonable approximation is created units correspond to buying while redeemed units correspond to selling. Accordingly, we can infer ETF investor activity at the aggregate level through the change in outstanding share count and price on each day.

For example, there is a popular fund, the KraneShares CSI China Internet ETF (ticker: KWEB), that invests in Chinese technology stocks. This sector has been in the news recently due to the Chinese government's action against many listed firms. But even after the recent sell-off, KWEB has doubled in price since inception on July 31, 2013.

Exhibit 1 shows KWEB's shares outstanding (left axis) and price (right axis). Looking only at days that KWEB shares were created (net buying) — the weighted average creation price of $54.10 can be estimated from the number of new units and respective closing prices. Similarly, for days that KWEB was redeemed, the weighted average redemption price is $50.15. Even as KWEB has gone up since launch, its investors have incurred an 8% average loss on transacted units by buying high and selling low. The average creation price is well above the average redemption price.

Often after a fund performs well, inflows arrive (and vice versa), particularly for thematic products. Hence, the average cost basis for ETF investors is routinely higher than the fund's average price as money follows past performance. For example, the ARK Innovation ETF (ticker: ARKK) is up about 500% since it opened in 2014.

Exhibit 2 shows ARKK's average price and average unit creation (purchase) price. Notably, the cost basis ($100) for the average investor in ARKK is far above the average price ($49) that the fund has traded at. This is because about half of ARKK units were created within the last year after its price had already surged.

Although this effect is most pronounced for thematic funds, it is observed in other products as well. The total return for the S&P 500 and the financial sector ETF is nearly identical over the past decade. However, flows into XLF have generally picked up after it has outperformed the S&P 500, and vice versa. Consequently, the average XLF investor has underperformed the S&P 500 as their market timing decisions have detracted value.

Investors are treating ETFs like stocks and using them to actively seek outperformance compared to a broad index. The total number of U.S.-listed ETFs is rising rapidly while the number of U.S.-listed stocks is falling due to corporate mergers and private equity buyouts. An already crowded marketplace encourages the launch of ever more niche funds.

The ETF structure can lull some investors into a false sense of security because it seems safer and "passive," but unless you are buying a slice of the entire market on a buy-and-hold basis, you are active. To be clear, the creation/redemption gap is not a fault of ETFs. It instead shows that ETF investors have the same tendency to chase returns as other investors.

Importantly, flows themselves also impact markets — when money is going into a sector, buying pressure pushes up prices of constituent stocks. And when a sector goes cold, selling pressure depresses prices too. It is often better to do the opposite — provide liquidity by buying when others need to sell and selling when others need to buy. Doing this has historically generated alpha over the long term, avoids crowding and reduces market impact.

Chasing a hot ETF is as active as chasing a hot stock. Mind the price gap between inflows and outflows regardless of the kind of investment vehicle you choose.

Aneet Chachra is a portfolio manager for multistrategy and global diversified risk premia strategies at Janus Henderson Investors, based in Los Angeles. This content represents the views of the author. It was submitted and edited under P&I guidelines but is not a product of P&I's editorial team.