SPDR ETFs: Liquidity and Price Discovery Tools

ETF spreads widen along with underlying constituents

Along with elevated trading volumes, bid-ask spreads increased globally across all asset classes and financial instruments amid challenging market conditions and liquidity constraints. ETFs around the globe experienced an increase in bid-ask spreads during March 2020 before reverting back toward their previous levels in April and May 2020.

In the US, increased volatility and liquidity constraints across underlying securities led to a temporary deterioration of traditional market quality metrics in ETFs.

• US-listed ETFs saw average bid-ask spreads increase from 0.37% in 2019 to 1.09% in March 2020. 
• SPDR ETFs saw their average bid-ask spreads increase from 0.14% in 2019 to 0.37% in March 2020.
   

Bid-ask spreads in ETFs widened, reflecting increased market risk and wider bid-ask spreads in underlying instruments.

For example, average spreads of constituents in the S&P 500 Index increased from 0.05% in January 2020 to 0.19% in March 2020. During that same time period, average spreads of U.S. Equity SPDR ETFs widened from 0.10% in January 2020 to 0.28%, reflecting the increased cost for market makers to transact in the underlying market. In some instances, ETF spreads traded within or less than those of their underlying constituents due to the following:

• Centralized Trading¹ Liquidity is centralized on exchange where buyers and sellers can transact on the secondary market. 
• Broader User Base ETFs tend to have two-way order flow due to their broad user base, as both investors and traders can utilize ETFs for a variety of reasons, including as a liquidity vehicle, hedging vehicle or simply to gain exposure to a specific asset class.
   

Price dislocation of equity ETFs

In addition to wider spreads experienced during the COVID-19 volatility, ETFs also traded at wider-than-normal premiums and discounts to Net Asset Value (NAV). Figure 3 indicates that during 2019, on average, U.S. Equity ETFs traded within a 1% premium or discount to their NAVs. However, in March 2020, ETFs experienced larger dislocations, which we can likely attribute to greater liquidity and market risk. That said, we believe the dislocations seen in U.S. Equity ETFs were further impacted by the following factors:

• Timing of the closing auctions of ETFs and stocks In the US, an ETF’s closing auction is conducted by its primary listing exchange at 4:00 p.m. Eastern Standard Time. Given the closing auctions of U.S. stocks don’t always take place at 4pm, particularly during March 2020, the official closing prices for stocks might not be available for market participants to use as inputs for ETF pricing. In fast-moving markets with large imbalances at the close, this can lead to larger-than-usual deviations between an ETF’s closing price and the NAV calculated with the closing prices of underlying constituents.
• Market volatility may trigger short sale restrictions Short sale restrictions on various single stocks in the US were triggered in March 2020 by Rule 201 of Regulation SHO. These restrictions can make it difficult for market makers to hedge risk using stock baskets, particularly into the close on those days in which this rule was in effect. This very likely contributed to larger-than-normal premiums/discounts. 
• Premiums/discounts may result from large imbalances during closing auctions. During the month of March 2020, ETFs and single stocks experienced significantly larger volumes and imbalances during closing auctions. These higher volumes at the close could have led to larger premiums/discounts, considering other transparency and hedging factors were also at play.

Market liquidity precipitated discounts on fixed income ETFs

Even more so, turmoil in fixed income markets led to significant dislocations between ETFs and their respective NAVs, resulting in larger-than-usual premiums/discounts.

These dislocations may have been attributed to the following factors:

• Stale or delayed NAV pricing Fixed income liquidity became challenged and pricing more opaque than usual. Fixed income ETFs, however, tend to reflect more real-time sentiment and realistic pricing levels as to where the basket of bonds should trade. As a result, pricing on individual bonds can lag behind the real-time market sentiment and executable pricing levels reflected by the ETF, resulting in the appearance of large discounts to NAV. In some cases, the ETF price may have been a better representation of actionable trade prices of the underlying constituents, when some were not always quoted by dealers, and thus acting as an efficient price discovery venue.
• Discrepancies between NAV strike and ETF closing times End-of-day NAV prices can be struck at a different time than the ETF’s closing auction, resulting in different values. Premium and discount are using only one point in time, thus not capturing intraday evolution, and many bond indices have different closing times, depending on the regions. Heightened intraday volatility in March exacerbated the impact of these timing differences. 

While some may question the performance of Fixed Income ETFs during March, we believe the discounts accurately reflected real-time market sentiment and liquidity of their underlying fixed income securities. Bond prices lagged real-time market sentiment and realistic trading levels. The surge in Fixed Income ETF trading volumes in March 2020 suggests that market participants gravitated toward ETFs as liquidity vehicles when individual bond liquidity became difficult. It is illustrated in Figure 5 — trading volumes of investment-grade and high yield corporate bond ETFs increased substantially as a percentage of cash bond trading during March 2020.

Price dislocation: Gold ETFs vs. Gold futures

In recent years, institutional investors have more widely adopted the use of ETFs as assets and liquidity have grown leading investors to more often compare structure, cost, and efficiency between ETFs and other investment vehicles.

A common comparison amongst certain institutional investors is ETFs vs futures. This relationship came to the forefront during several days in March, when futures and ETFs that track the spot price of gold diverged.

On March 23, 2020, three major Swiss gold refineries in the world announced that they were suspending production entirely for at least a week due to the outbreak of COVID-19. With flights grounded and gold refineries closing, traders became concerned that it would become challenging to meet futures contract settlement obligations if they were unable to move 100-oz COMEX bars from Europe to New York.

As a result of supply chain concerns, on March 24, 2020, with the demand for gold still apparent due to the COVID-19 crisis, gold futures in New York and gold spot prices in London diverged significantly. The gold futures premium extended over several weeks in March and April 2020.

Figure 6 indicates that this premium rippled through the gold ETF market, with the largest gold ETF listed in the US trading at a premium on March 24, 2020. However, this ETF resumed trading closer to its fair value after only one day, on March 25, 2020. This is because this particular ETF tracks the gold spot market, which operates primarily out of the London gold market hub and continued without any reported issues.

The structure of this ETF played an important differentiating factor for investors seeking to track the gold spot price during this period of market stress.

In summary, despite some of the aforementioned market quality and price dislocations during the COVID-19 pandemic, ETFs have provided investors with liquidity when they needed it most. Read more about the road ahead.