From the graph below, it is evident that, while investors do express investment views on the high yield asset class, even in extreme cases, such views tend not to dramatically affect the liquidity of the underlying market.
Indeed Zarate points out: “The bulk of operations is regularly netted off between buyers and sellers of existing ETF shares, while, whenever needed, creation and redemption activity in the primary market has not presented problems for execution.”
The graph above looks at the US market, which is easier to analyse because trading on the exchange is the norm and the iShares iBoxx $ High Yield Corporate Bond ETF (HYG) and the SPDR Barclays High Yield Bond ETF (JNK) account for around 65% of all AUM held by US-domiciled high-yield ETFs.
However, Zarate says, in the case of European-domiciled ETFs where it is estimated that up to 70% of secondary market trading takes part away from the exchange, “impact of ETF trading on the liquidity of the underlying high-yield bond market is limited”.
“In fact, we find that secondary/primary market ratios have tended to spike considerably at times of stress in the high-yield bond market, while remaining within range at times of above-average redemptions from the ETFs,” he said.
According to the report, while there are plenty of periods when the secondary/primary market ratio spikes significantly above median values, which is an indication the impact of trading on the underlying, in many cases “these spikes coincide with periods of stress in the underlying–for example, the start of the eurozone debt crisis in 2009, the US Fed taper tantrum in the summer of 2013, or the comprehensive sell-off in late 2015.”
But, it adds: “Even in episodes of heavy redemptions from high-yield bond ETFs, such as the one experienced in early May 2015, we see that secondary market activity continued to greatly outpace the net impact on the primary market. Indeed, the ETFs behaved as expected, and the flow of redemption orders was duly executed.