The Analytics Group of IIROC performed a Trading Review and Analysis of High Frequency Trading on Canadian equity markets. IIROC uses a methodology to identify user IDs exhibiting high order-to-trade ratios, or HOT User IDs, and covers the period from August 1, 2011 to October 31, 2011. With User ID data provided by the Regulatory feed IIROC determined HOT User ID’s by isolating the actions of placing an order, amending that order, then having the order trade. Canceling an order and sending a new order is isolated against sending an order and amending that order.
Plotting the number of trades against the number of new and amended orders yields:
After using their methodology, HOT User IDs are identified in red with OTHER User IDs in blue:
So just who are these Hot Users? They are:
- 11% of the User IDs active in the study period;
- 22% of share volume traded;
- 32% of dollar value traded;
- 42% of trades executed; and
- 94% of new and amended orders.
The HOT Users encompass the group we would expect, those with fast processors, the fastest pipes, and the primary data feeds. This group has been identified using the Regulatory feed account types as either:
- DMA (Direct Market Access)
- NDMA-IN (Non-DMA Inventory) – Inventory meaning activity related to proprietary trading activity
- NDMA-CL (Non-DMA Client)
- NDMA-Other (ie) specialist, non-client, options market-maker.
Ploting the order to trade ratio of Hot User IDs by account type yields:
As I highlighted in a post on Floating Path in June of 2012, ETFs are dominated by HFT over all other categories. IIROCs study reached a similar conclusion regarding the level of HFT activity in ETFs. This is another aspect of what is perverted within the current market structure: the derivatives now price the underlyings as it is easier to calculate in rapid fashion, a group of set prices, such as a SPYs constituency, instead of the fundamental analysis required to price the individual companies. This rapid calculation leads to repeated revisions and makes the ETF market more appealing for those looking to execute trades on the predictive power of price movement over the predictive power of a public company’s earnings. The Canadian market is populated in much the same fashion, with the NDMA-Other – specialist, non-client, options market-maker account type having the largest ratio:
Whether measured by volume traded, valued (sic) traded or number of trades executed NDMA-CL and DMA were more active in common shares and NDMA-IN and NDMA-Other were more active in ETF/Ns.
With the individual activity identified, we can now measure the interaction broken down by order types and whom they traded with:
27% of the volume by Retail participants involved interaction with HOT Users, as you’d expected, being the DMAs. As for the dark markets in Canada, HOT Users were responsible for a larger percentage of activity than on the lit markets, 39% vs. 21%:
The paper continues into an analysis of rebates earned, security types, liquidity categories, price bands popularity levels, inter-listed security activity, New and Amended Orders and Order-to-trade ratios, and trading venues. The paper is must read for all those interested in identifying and valuing the activity levels of specific types of participants in equity markets. It is one of the more transparent papers and it draws on input from many of the papers included in R.T. Leuchtkafer bibliography.