India’s Equity Market Microstructure Evolution
India’s equity market microstructure has undergone a significant transformation over the past few years. The key drivers of this change have largely been regulatory as Direct Market Access (DMA) and Smart Order Routing (SOR) have been permitted by the Security and Exchange Board of India (SEBI). Technology has also been a factor as both the National Stock Exchange of India (NSE) and the Bombay Stock Exchange (BSE) have upgraded the exchange matching engine, improved price dissemination and introduced collocation. Both the regulatory changes and technical enhancements have also ushered in increasingly sophisticated trading systems and algorithmic trading that have seen orders increase, execution ratios decrease, order sizes decrease and spreads tighten, though with not much liquidity depth in the order book, to around 3bps. Through two time series videos produce by Deutsche Bank we will show you India’s market microstructure changes and attempt to explain them.
Video 1 is a time series of how the microstructure has changed starting from June 2007. The BSE is on the left and the NSE on the right. On the X-Axis is the total notional posted in millions of rupee and on the Y-Axis is the spread in basis points. The size of the circles represent volatility. As we can see in June 2007 the NSE has somewhat more liquidity than the BSE with a similar volatility profile. Spreads however, are much tighter on the NSE averaging around 4-5 bps with the BSE showing around 9- 10 bps.
At the end of 2007 we see volatility pick up particularly on the NSE as the NIFTY reached an all time high of 6138 in November. The spreads on the NSE moved up to around 7bps and the amount posted to the order book almost doubles. As the selloff in early 2008 gains momentum the BSE beings to exhibit increasing volatility and order book notional. In April 2008, SEBI mandates DMA and we can see, despite the increasing momentum of the Global Financial Crisis (GFC) that order book on the NSE displays much more liquidity. The BSE eventually exhibits a similar behavior but is overshadowed by the Lehman collapse in September 2008. As there is more liquidity on the NSE traders prefer to access this exchange first causing connectivity and access to lag on the BSE. As some sense of normalcy returns to the markets in late 2009 we can see spreads tighten to around 3-4 bps on the NSE and around 6-7 on the BSE. The NSE is showing increasing liquidity as well. The NIFTY tests its old highs in late 2010 but we can see much lower volatility and spreads compress to 2-3 bps on the NSE and around 5-6bps on the BSE with a few names averaging 3bps.
The impact of SOR is yet to be felt as this was only permitted in August 2010 with the first broker, Deutsche Bank, being approved only in November of 2010. At the time of this writing there are 11 brokers approved to offer SOR in India.
Video 2 is also a time series of quotes vs trades with the BSE again on the left and the NSE on the right. The X-Axis shows number of quote changes per day and the Y-Axis shows the number of trades per day. In June 2007, we can see that the ratio of trades to quotes is 5 to 1 for the NSE over the BSE. This continues to the end of 2007 when the NIFTY was making its all time high up to around April 2008 when DMA is finally allowed in India. We can then start to see the BSE increasing both its quotes and orders per day. This becomes more pronounced as the GFC is at its peak early 2009. Then, dramatically, in June 2009 the NSE sees a large increase in quotes and executions per day. The BSE in September 2009 follows suit. Again, the BSE lags investment in connectivity as the industry would rather connect to the NSE first as it has the most liquidity. Then, in May 2010, we see the BSE liquidity profile change dramatically as the number of quotes increases substantially. This is indicative of automated market making. The BSE has no restriction on how many times you can update quotes in a day but the NSE does when the order to trade ratio exceeds 500 to 1 making the BSE better suited to HFT (but without the liquidity) when updating quotes constantly. As mentioned above SOR impact is not seen in this time series video.
In conclusion, the uptake of DMA, technical improvements at the exchange, collocation and the uptake of automated trading have had a significant impact on the equity market microstructure in India. While the impact of SOR has yet to be felt and the costs to implement present a high barrier to entry, India’s electronic trading industry is certainly evolving. It has the highest fragmentation of any of Asia’s markets (though there are only 2 venues), by far the lowest spreads in Asia and an exchange infrastructure to rival any in the west.