Chinese Markets are Closed--So What Happens to China-based ETFs in the US?

Happy Chinese New Year! Markets were closed in many Asian countries last week, while US markets remained open. As noted by several commentators, this means that while US ETFs that hold Chinese equities were actively traded, their underlying assets were not. So what does this mean for China-based ETFs traded in the US?

First, it's important to note how ETFs relate to their underlying assets. Essentially, ETF shares can be created by certain traders (called authorized participants) by buying the underlying assets of the ETF (in this case, shares of Chinese stocks) and delivering them to the ETF issuer. This process allows the price of the ETF to stay 'in line' with the price of the underlying assets.

However, if the authorized participants cannot buy the underlying stocks, this price discovery mechanism cannot take place. This is actually not a rare occurrence when it comes to Asian or Australian ETFs: because these markets are open when US markets are closed overnight and vice versa, this actually happens every day. What happens is that active participants make guesses as to the value of the underlying securities. From iShares:

Now, this may seem like a hassle for the authorized participants, but the truth is there are many ways for them to make educated guesses about where these securities should trade when their exchanges are closed. One popular method is to use the percent change of the US market (for example, via the S&P 500) and assume that other countries will have a similar move, while also incorporating the continuously traded foreign exchange market to account for currency movements.

There are two ways to look at this issue: either the US ETF market is providing price discovery to assets whose native markets are closed, or the US ETF market is deviating from the value of the underlying assets during the trading day. While this is to some degree a matter of perspective, we decided to look at the largest ETF holding only Chinese equities (FXI) and compare it to the largest holder of US equities (SPY). Specifically, we wanted to see whether overnight moves were larger for FXI or for SPY on average. If so, that would suggest that price discovery was happening overnight (when Chinese markets are open) rather than during the US trading day.

We found that between 2005 and 2013 the mean absolute overnight return was much larger than the mean absolute intraday return for FXI. The opposite is true for SPY. This suggests that larger moves happen overnight for FXI and intraday for SPY. We also tested the median absolute return, the standard deviation of absolute returns and the standard deviation of returns which all showed the same pattern.

It's important to understand how ETFs work and how their returns relate to the value of the underlying assets. In this particular case, it seems that although FXI trades on a US exchange the majority of the price discovery may occur overnight rather than intraday like most US equities.