Economic impact assessments of proposals for computer trading

Sector:Finance | Financial markets and Institutions
Published: September, 2012
Document type:Report 
Tagged: cost-benefit analysis impact assessment quantitative analysis UK

The report provides an evidence-based analysis of a number of proposals targeted at computer trading within financial markets, with some of these measures having been included in the European Commission’s proposal for a revised Directive on markets in financial instruments (MiFID II). These measures include: increasing tick sizes and introducing control by regulators of algorithms, among others.

The impact assessments follow the EC Impact Assessment Guidelines. In the present context this entails assessing the wider economic impacts – on the cost of capital, the overall level of financial intermediation and economic activity – of changes to financial market microstructure that would come about as a result of introducing the measures under consideration.

There are a number of key results relating to each of the measures assessed, detailed in the main report, including (for those above mentioned proposals):

i. Increasing tick sizes are expected to bring about a widening of bid-ask spreads and associated, real economic costs. Quantitatively, a reversal of the tick size reductions piloted on European trading venues in 2009 (including, BATS and Chi-X) are estimated to bring about a cumulative loss in EU27 GDP (at constant prices) over a period of 10 years of €60bn. This is despite the notion that increasing tick sizes could improve inter alia liquidity.

ii. The aim of regulatory control of algorithms is to ensure that no “rogue” algorithm could create havoc in financial markets. If effective, this would reduce the risk faced by liquidity providers and hence reduce bid-ask spreads relative to a situation where such a risk of rogue algorithms is very real. Stakeholders all agreed that it would not be possible for regulators to undertake in-depth reviews and analysis of the many different algorithms being used. However, many stakeholders were of the view that a more robust regulatory approach to ensuring that algorithmic trading firms properly test the robustness and resilience of their algorithms (with a clear audit trail of the testing and assessment undertaken) would be feasible and beneficial.   Based on an analysis of developments of bid-ask spreads around the so-called flash crash, the impact of regulatory control of algorithms through the approach put forward by stakeholders is likely to be small, boosting EU27 GDP (at constant prices) in the long run in the order of 0.1%.