Dark pools, such as Turquoise Plato (previously Turquoise Midpoint Dark), are trading venues that do not offer pre-trade transparency. This means that one can submit an order to such an exchange without anyone being the wiser until the order has been executed. The debate on the usefulness or the market quality imperativeness of these type of venues is a long-running one. Stakeholders are especially concerned about the effects of the lack of pre-trade transparency. In 2010, the US Senator Kaufman, in a letter to the then SEC Chair Schapiro, calls for actions to “examine whether too much order flow is being shielded from the lit markets by dark venues”. A survey conducted among market participants by the CFA Institute found that 71% of respondents are of the view that dark pools constitute problems for price discovery in the markets (see Schacht et al. 2009). At first glance, the scepticism appears valid. A lack of pre-trade transparency suggests that dark pools impair price discovery, and this perception has led to efforts to increase disclosure and curb trading in dark pools in several countries, including Australia, Canada (through price improvement rules), and in Europe (through the imposition of volume caps).
At the heart of the impact of dark pools on market quality are the dynamics of trading venue selection in financial markets. Understanding these dynamics, and specifically, what drives trading volume to dark pools, is critical for the determination of the effects of dark trading on market quality. This need has never been more critical than now as dark trading currently enjoys high prominence in developed markets. European dark pools accounted for 9.1% of all European on-exchange activity in April 2019 and for 9.6% in July 2019. Note that in 2019 caps had already been placed on dark trading under the EU’s Market in Financial Instruments Directive (MiFID) II (see here and here). The influence of dark pools is even larger in the US, where dark pools and other off-exchange trading venues executed 38.6% of US equity volume in April 2019. Although market quality is just a means to an end, it is nonetheless very important. How well financial markets perform their key functions of providing liquidity to facilitate hedging, diversification, and saving, and efficient price discovery to direct resources to their best uses within the economy, affects every individual’s financial well-being. For example, a major fraction of all retirement savings is invested in global stock markets either directly or indirectly. Job growth, incomes, and therefore our standard of living depend on corporate investment, the financing of which in turn depends on well-functioning financial markets. Therefore, when markets experience the kind of turbulence observed this Spring, we should pay attention.
Theory tells us that dark trading dynamics are driven by price volatility in lit exchanges (see Zhu 2014); lit exchanges, such as the London Stock Exchange’s main order book, the Stock Exchange Electronic Trading System (SETS), are transparent and allow all market participants to observe every order submitted to them prior to their execution. However, the endogenous determination1 of volatility makes it challenging to test this prediction. In a recent paper (see Ibikunle and Rzayev 2020), we avoid this problem by exploiting the novel COVID-19-induced market volatility observed in late February and most of March 2020 (see Baldwin and Weder di Mauro 2020, Baker et al. 2020) as an exogenous shock to investigate volatility’s role in venue selection by informed and uninformed traders in today’s financial markets. This also allows us to capture the dynamics in dark pools’ market share and volume as the COVID-19-induced market volatility evolves, and ultimately, to analyse the implications of the event for market quality.
According to Zhu (2014), when a dark pool is added to a market with only a lit exchange, traders that are not well-informed about the value of the security they are trading (called uninformed/liquidity traders) are likely to move their trading activities to the dark pool. This is because they are wary of better-informed traders (called informed traders) that already trade on the lit exchange. The informed traders will stay on the lit exchange as long as price volatility is not too high, and the cost of trading or the risk of not executing their orders is minimal. As informed trader concentration increases gradually on the lit exchange due to uninformed traders leaving for the dark pool, volatility widens the exchange spread (the gap between buyer and seller prices) and encourages more uninformed (liquidity) traders to migrate to the dark pool – this is the natural state of things when volatility is moderate. Informed traders stay at the lit exchange, because when volatility is at a moderate level, the exchange spread is not excessive and thus, non-execution risk in the dark pool is higher than any potential price improvements they may offer (for example, in Australia and Canada, price improvement is required to trade in dark pools). However, when volatility in the exchange exceeds the maximum level needed for informed traders to avoid the dark pool, informed traders start to migrate to the dark pool in search of uninformed counterparties to trade with and in a bid to avoid the widening exchange spread. Thus, liquidity constraints in the lit market can result in informed traders entering non-transparent/dark venues in order to reduce their transaction costs and increase their profits (Hendershott and Mendelson 2000, Nimalendran and Ray 2014). The informed traders’ migration consequently results in uninformed traders leaving the erstwhile safety of the dark pool for the lit exchange.
The results we obtain in Ibikunle and Rzayev (2020) strongly support the theoretical predictions outlined above. We exploit both the excessive exogenous market volatility induced by the spread of COVID-19 and the MiFID II double volume cap dark trading restrictions currently in force in some European stocks, to investigate the role of volatility in the evolution of dark market share and the decision of where to trade for both informed and uninformed traders. Our sample includes 55 of the stocks with dark trading restrictions (control group), as well as an additional group of 55 stocks with no dark trading restrictions (treated group). Analysing and comparing the dark trading and lit exchange trading dynamics in these 110 stocks then allows us to examine the effects of exogenous volatility on informed and uninformed traders’ trading dynamics when they can trade in both dark pools and lit exchanges. It also affords us an opportunity to measure how the evolution of volatility drives market quality.
Our results show that the excessive volatility observed from 24 February 2020 at lit exchanges (see Figure 1) lead to an economically significant shift of informed trading activity from lit venues to dark pools.
Figure 1 Volatility
Note: The figure plots the day-by-day evolution of the cross-sectional average of Volatilityi,d for 110 European stocks employed in the study. Volatilityi,d is computed as the standard deviation of hourly mid-price returns for stock i on day d. The sample period covers 24 January to 24 March 2020. The vertical bar indicates 24 February 2020, when the COVID-19-induced excessive volatility is adjudged to have commenced in global financial markets.
We also show that this shift of informed traders in turn drives the migration of uninformed traders, who are, wary of being adversely selected, moving from dark pools to lit venues. The net effect of the cross-migration is a loss of market share by dark pools and an increase in lit venues’ market share as the share of uninformed traders in financial markets is almost always larger than the share of informed traders. Figure 2 shows that during the more volatile trading period beginning on the 24 February 2020, the market share of dark pools declines from 2.5% during the period preceding the market volatility phase to 2.1% for the period between 24 February and 24 March 2020.
Figure 2 Dark trading
Note: The figure plots the day-by-day evolution of the dark volume and dark market share for 55 European stocks that could be traded at both lit and dark venues. Dark market share is computed as the dark trading volume for a given day divided by the total trading volume on the same day. The sample period covers 24 January to 24 March 2020. The vertical bar indicates the 24 February 2020, when the COVID-19-induced excessive volatility is adjudged to have commenced in global financial markets.
Although dark trading volume in the treated stocks doubles during the excessive volatility period, this only reflects the overall increase in trading activity due to the COVID-19 pandemic (Figure 3). This implies that the increase in trading activity, which is inevitable given the urgency created by the spread of the COVID-19 pandemic, is higher in the lit venue. The difference between the two periods is both statistically and economically significant.
Figure 3 Trading volume
Note: The figure presents the day-by-day evolution of lit volume for 110 European stocks. Panel A presents the day-by-day evolution of lit volume for the full sample (both the 55 stocks that could be traded at both lit and dark venues, i.e. treated stocks, and the 55 stocks with dark venue restrictions, i.e. control stocks), while Panel B shows the day-by-day evolution of lit volume for the control and treated groups separately. The sample period covers 24 January to 24 March 2020. The vertical bar indicates the 24 February 2020, when the COVID-19-induced excessive volatility is adjudged to have commenced in global financial markets.
These dynamics have mixed implications for market quality characteristics. Although the market liquidity of lit exchanges improves as spreads narrow during the volatile trading period for stocks that have no dark trading restrictions compared to those with trading restrictions, price discovery deteriorates for the former group of stocks in comparison to the latter because informed traders migrate to the dark pools.
Thus, it appears that volatility is a market-regulating mechanism driving the share of trading activity in dark pools, which has implications for market quality and welfare. Regulators should account for this when designing regulatory mechanisms, such as dark trading caps and waivers. Interventions need to be flexible enough to account for changes in market conditions, such as periods of exogenously-driven high volatility as provisions designed for normal trading conditions become irrelevant when markets are impacted by events such as a rampaging virus.
Baldwin, R, B Weder di Mauro (2020), Mitigating the COVID economic crisis: Act fast and do whatever it takes, A VoxEU.org eBook, CEPR press.
Baker, S R, N Bloom, S J Davis, K Kost, M Sammon, T Viratyosin (2020) “The unprecedented stock market reaction to COVID-19”, Covid Economics: Vetted and Real Time Papers 1, 3 April.
Hendershott, T, H Mendelson (2000) “Crossing Networks and Dealer Markets: Competition and Performance”, The Journal of Finance 55: 2071-2115.
Ibikunle, G, K Rzayev (2020), “Volatility, dark trading and market quality: evidence from the 2020 COVID-19 pandemic-driven market volatility”, Covid Economics: Vetted and Real Time Papers, forthcoming.
Nimalendran, M, S Ray (2014), “Informational linkages between dark and lit trading venues”, Journal of Financial Markets 17: 230-261.
Schacht, K, C Cronin, C J Allen, R Preece (2009), Market Microstructure: The Impact of Fragmentation under the Markets in Financial Instruments Directive, Charlottesville, VA: CFA Institute Centre for Financial Market Integrity.
Zhu, H (2014), “Do dark pools harm price discovery?”, The Review of Financial Studies 27: 747-78.
1 This implies that although volatility affects dark trading, dark trading also influences volatility, and this makes it difficult to disentangle the effects of one from the other. The exogenous event we use in this study is driven by the spread of a virus that arguably has no comprehension of modern market structures nor directly responds to them.