Topics in Market Microstructure

The increase in trading volume raised concerns about the impact of algorithmic trading, which includes high frequency trading (HFT), on price discovery process and volatility. First chapter considers the regulatory debate concerning HFT which led the European Commission to suggest implementing a financial transaction tax (FTT) of 0.1% on all stock transactions. A simulation of pure electronic limit order book (E-LOB) finds support to implement FTT. Particularly, traders are found to trade more aggressively by increasing the volume traded to remain in a profitable position after tax. The market, nevertheless, ended up with higher trading volume, lower bid-ask spread and almost same price volatility. The second and third chapters rebuild the London Stock Exchange Electronic Order Book (SETS) in real-time for 5 different stocks for two consecutive months, July and August 2007 and utilise the Directional Changes (DC) methodology to track price trends. Notional Volume Weighted Average Price (NVWAP) concept is used to analyse the intraday dynamics of liquidity in the London Stock Exchange E-LOB; namely, the slopes of NVWAP curves and the volumes on both sides of the market (bid and ask) are studied. Second chapter observes that the shape of both sides of the order book changes during DC in predictable ways where changes in volumes and NVWAP curves’ slopes revealed to be robust proxy to identify the prevailing market trends without prior knowledge of price. Third chapter assesses order book events’ (OBE), i.e. submissions and cancellations, influence on the shape of the order book which generates price trends that may cause flash and mini-crashes. It revealed that OBE’ effects are highly significant determinants of the change in cumulative return under normal price conditions while only buy side events are significant under extreme price conditions. These findings match the expected direction of change in cumulative return.