Ever since we heard that some of the largest ETF issuers were lining up emergency liquidity lines to tap in the event all the retail money that’s piled into things like HY debt suddenly decides to head for the exits, we’ve gone out of our way to explain just why it is that the likes of Vanguard would consider paying out redemptions with borrowed money as opposed to selling the underlying assets. 

The problem is that in the context of the post-crisis regulatory regime, banks are no longer willing to hold large inventories of securities and so, when a bond fund manager facing large outflows tries to transact in size, he or she will likely be confronted with an extremely thin secondary market. Rather than risk dumping a large amount of assets into a market with few buyers and thus facilitating a fire sale atmosphere, fund managers are considering paying out investors who sell with borrowed cash and selling off the underlying assets slowly as the market permits. 

ETFs and other portfolio products mask this problem as long as flows are diversifiable. That is, if fund manager A is experiencing outflows, that’s ok as long as portfolio manager B is seeing inflows. However, once flows become unidirectional (i.e. everyone is selling), then managers will need to go and sell the underlying assets and that, in today’s market, is a big problem. Thus, ETFs give the illusion of liquidity - that is, investors assume there’s no problem because they can trade in and out easily, but should the flows suddenly all head in the same direction everyone will quickly discover that, as Howard Marks put it, an ETF "can’t be more liquid than the underlying."

Amusingly, a new academic study from researchers at Stanford, UCLA, and the Arison School of Business in Israel suggests that ETFs themselves are contributing to a lack of liquidity for the stocks they hold. Essentially, the argument is that increased ETF ownership leads to wider bid-asks, less analyst coverage, and higher correlations with broad market moves.

Specifically, the hypotheses the researchers tested were: "1) An increase in ETF ownership is associated with higher trading costs for the underlying component securities, and 2) An increase in ETF ownership is associated with a deterioration in the pricing efficiency of the underlying securities."

On the first hypothesis, the authors looked at "the relation between ETF ownership and two proxies of liquidity that capture trading costs: (1) bid-ask spreads, and (2) price impact of trades." On the second, they looked at "the extent to which stock prices reflect firm-specific information." 

The results: 

We first demonstrate that an increase in ETF ownership is associated with an increase in firms’ trading costs. This is consistent with the idea of uninformed traders exiting the market of the underlying security in favor of the ETF. As uninformed traders exit and trading costs rise, we posit that pricing efficiency will decline. Consistent with this prediction, we find that higher levels of ETF ownership are associated with an increase in stock return synchronicity and a reduction in future earnings response coefficients. We also observe a negative association between ETF ownership and the number of analysts covering the firm. Collectively, the evidence presented in this paper suggests increased ETF ownership can lead to weaker information environments for the underlying firms. 

And here’s WSJ summarizing:

ETFs divert many individual and institutional investors from trading directly in certain stocks. And that means fewer opportunities for professional traders to profit from trading those stocks. Ownership of U.S. stocks by ETFs has grown from 0.1% of shares outstanding in 2000 to 7% in 2014, according to the paper.

The diminished profit potential leads to less competition for shares among traders, and that, in turn drives up the cost of trading the shares for everyone, as measured by the so-called bid-ask spread, the authors say. This spread is the difference between the price a stock can be sold at on the market and the (higher) price it can be bought for at any given moment.

The bid-ask spread is 6% wider on average when a big chunk of a company’s shares—more than 3% of the shares outstanding—is held by ETFs, compared with the spread when a stock has lower or no ownership by ETFs, the study found.

Wider bid-ask spreads not only increase the costs of trading, they also further reduce profit potential for traders. That gives traders less incentive to bid for stocks in anticipation of future earnings increases at the issuing companies. So stocks become less responsive to projected earnings, the report says.

There also is less financial incentive for analysts to provide company-specific analysis for stocks with less profit potential. The study found that the number of analysts covering a stock falls as ETF ownership of the company increases.

Finally, less company-specific analysis also means that a greater proportion of a stock’s price moves are likely to be driven by industrywide or general market movements.

Obviously, some of this is self-evident, but the important thing to note is that this looks to be still more evidence of a wholesale shift away from trading underlying asset classes in favor of trading derivatives.

And while we might be able to distinguish between those who are intentionally avoiding the underlying markets due to perceived illiquidty (i.e. fund managers trading portfolio products to meet redemptions and satisfy inflows and traders resorting to futures to avoid illiquid cash Treasury markets) and those who are simply not trading the underlying because trading the alternative is easier (i.e. retail investors opting for ETFs over invdividual stocks because they don't feel they have the "sophistication" to trade the individual names) the effect is the same: the market for the underlying assets becomes ever more illiquid. 

Etf Paper