- The gap between an ETF’s price and the net asset value (NAV) of the fund can be an important source of risk for investors.
- For many equity ETFs that hold a large fraction of their constituents’ shares, the maximum price-NAV difference is close to transaction-cost estimates built from MSCI’s RiskMetrics® LiquidityMetrics® and historical fund flows.
- Transaction costs may provide a forward-looking view of price-NAV risk, which may enable investors to more effectively manage this source of risk.
As thematic equity ETFs have grown in popularity, concerns have been raised about their liquidity — especially when an ETF holds a large fraction of a constituent’s shares.1 In particular, because of the mechanics of ETFs, investors are free to transact intraday in the fund, but the underlying equities may struggle to keep up with the fund’s trading activity. This liquidity mismatch could lead to a gap between the ETF price and the net asset value (NAV) of the fund. This gap can also be sizeable: For some funds the price-NAV difference reached 8%-10% during the COVID-19 crisis, while transaction-cost estimates reached 1%-2%. How can ETF investors get a handle on this risk?
Transaction Costs and Price-NAV Risk
One approach can be to look at historical price-NAV differences, but this is backward-looking and may not anticipate sudden market changes. Alternatively, investors may look at transaction costs. The ETF price stays close to the NAV through an arbitrage mechanism where market players (authorized participants, or APs) can profit from, and quickly eliminate, price-NAV discrepancies. While the details of this mechanism are highly complex and specialized, a simple model would say that the price-NAV should never be higher than the transaction costs that the AP would incur to eliminate the arbitrage. If it were, it would be a clear arbitrage for the AP.
We study a sample of thematic equity ETFs that have between USD 1 and 20 billion in assets under management (AUM), and also hold a significant fraction of their constituents’ shares. We estimate transaction costs for these ETFs using MSCI’s RiskMetrics® LiquidityMetrics® and historical fund-flow data. For most of the funds studied, these transaction-cost estimates were close to the maximum price-NAV difference during our sample period. These findings may suggest that a transaction-cost model could be used by ETF investors as a forward-looking gauge of the price-NAV risk of their ETF holdings.
Liquidity and ETF Prices for Concentrated Funds
To identify funds that hold a significant fraction of their constituents’ shares, we compute the number of shares in the ETF relative to the total shares outstanding, which we call the ownership ratio. We then rank ETFs by their weighted average ownership ratio and select the top nine for our analysis.2
The transaction costs depend on both the liquidity of the underlying shares and on the transaction size. In order to define a simple and conservative estimate of transaction costs, we set the transaction size to be the 99th percentile of historical daily fund flows over the sample period. The characteristics of these nine funds are shown in the table below.
The exhibit displays the historical price-NAV differences alongside our model transaction-cost estimates. For many of the funds, this simple model appears to provide a reasonable estimate of the maximum observed price-NAV differences. One exception is ETF 5, for which the transaction-cost estimate is far above the realized price-NAV difference most of the time. For this fund, it is possible that the 99th percentile of fund flows is over-conservative and a smaller transaction size would be more representative of an AP’s expectations.
Transaction-Cost Estimates Versus the ETF Premium/Discount
Select an ETF from the dropdown. Use the slider on the left to select different domains of the chart. Zoom in further using the mouse scroll on the slider. Source: IHS Markit, MSCI
We note also that on a small number of days, the observed price-NAV difference falls outside of our transaction-cost estimates. In these cases it is possible that the AP’s flow expectations were greater than the 99th percentile of historical flows. Other assumptions may also have contributed to these outliers, such as same-day liquidation of the basket of equities. In addition, for global ETFs, the price-NAV difference can be driven by different market-closing times.3
Characteristics of the Sample of Studied ETFs
| Weighted Average Ownership Ratio | AUM | Average Holding Size | 99th Percentile of Flows | |
| ETF 1 | 10.0% | USD 1-5 billion | Between USD 50-100 million | 28.2% |
| ETF 2 | 5.9% | USD 5-10 billion | Between USD 100-300 million | 1.4% |
| ETF 3 | 4.5% | USD 10-20 billion | Between USD 100-300 million | 1.3% |
| ETF 4 | 4.1% | USD 5-10 billion | Between USD 50-100 million | 4.3% |
| ETF 5 | 3.7% | USD 5-10 billion | Between USD 100-300 million | 3.6% |
| ETF 6 | 3.1% | USD 1-5 billion | Between USD 50-100 million | 3.4% |
| ETF 7 | 3.1% | USD 5-10 billion | Between USD 100-300 million | 5.5% |
| ETF 8 | 3.0% | USD 1-5 billion | Between USD 50-100 million | 1.3% |
| ETF 9 | 3.0% | USD 1-5 billion | Between USD 50-100 million | 5.9% |
Weighted average ownership ratio is the percentage of shares held by the ETF to the number of shares outstanding. Data as of April 15, 2021.
Looking at Costs to Estimate Risk
Even though the ETF price remains close to the NAV most of the time, the difference between price and NAV is a source of risk for ETF investors, especially when market liquidity dries up. Transaction costs may provide a forward-looking view of price-NAV risk, which may enable investors to more effectively manage this source of risk.
1Boyde, Emma. 2021. “Concerns grow over ETFs’ illiquid holdings.” Financial Times, March 29, 2021.
2Aggregation is done using fund weights in each underlying constituent.
3In our sample, ETF 3, ETF 4 and ETF 7 are global ETFs.