LMAX Exchange operates several order books that fall within the scope of MIFID II. We will frequently refer to the 'unit' and 'currency' of an order book.
Let's consider the UK 100 order book as an example. The currency of this order book is GBP. This means that any trading that occurs in this order book will have a cost in GBP; participants pay (or are paid) GBP for 'units' of the order book. The unit of this orderbook is some fraction of the CFD's underlying index determined by the contract size.
For each order book, detail is given on the currency and unit in Table 2.
We will at times refer to "an order book action". Instructions reaching an order book trigger a single 'action'. Before and after the action the state of the order book is considered to be a coherent market state, at which market data could be published. Examples of instructions would be placing an order, cancelling an order, placing a mass order, etc.
Several fields refer to the concept of an "accepted" order. Before considering an order for matching, the LMAX Exchange applies a number of rules to test the validity of an order (e.g is the price valid? is the order book open?). Only once all of this validation has succeeded is an order accepted.
On LMAX Exchange, participants are classified as either a market maker or a market taker. Market makers, in general, are expected to provide a two-way price in a number of order books. Market takers, on the other hand, usually only enter the market when they wish to trade. In certain tables we have separated some columns as defined in the original RTS 27 report into maker and taker equivalents to ease understanding and meet the spirit of the regulation.
Market makers typically have two ways of interacting with an LMAX Exchange order book.
The first is via "mass orders". Here, a market maker specifies a list of orders that they would like to place on to an order book. On receipt of a mass order, LMAX Exchange:
In the presented tables, this is treated as cancellation rather than modification
The second mode of interaction is via "cancel and replace". In this mode, market makers communicate each order they wish to place individually. Each order instruction is associated with a unique instruction id. Later on, if the market maker wishes to change the price of this order, they can send a separate "cancel and replace" instruction, referencing the original instruction by its ID.
In the presented tables, this is treated as modification, rather than cancellation.
Daily files are in a nested folder structure. For example, Table 8 for the 4th of March will be found in
/table3/year=2018/month=03/day=04/table-8.csv
Quarterly files can be found at the root of the export, so Table 2 and the quarterly element of Table 1 will be found in
/table2.csv and /table1.csv
Files are published per UTC day, excluding Saturdays which are non-working days.
Table 1 has been split into two components.
The first part will be published once per quarter, as it is constant:
The country of competent authority
The name of the market segment
The ISO 10383 market segment MIC
The second part of Table 1 will be published for each trading day.
The number of outages
Semicolon delimited list of the nature of outages. May be blank in the case of no outages.
The average outage duration, or "N/A" if no outages occurred
The number of trades cancelled
The percentage of total transaction value that was cancelled, to the nearest 0.1%.
Table 2 will be published quarterly, as the order books it pertains to are constant. The following fields will be included:
The name of the order book.
The ISIN of the order book.
The CFI code of the order book.
The currency of the order book.
The unit of the order book.
Table 3 is provided in a file per order book per day.
The file is named by the order book is represents.
The time point of this record. The record refers a two minute window beginning at this time.
The size range of this record.
"1" denotes smaller than the size specific to the instrument (EUR 90000).
"2" denotes between the size specific to the instrument (EUR 90000) and large in scale (EUR 100000).
"3" denotes large in scale (larger than EUR 100000).
The average executed price in the specified two minute window, denominated in the currency of the order book.
The total value of executions in the window, denominated in the currency of the order book.
The price of the first order in this size range that occurred after this window, if no executions occurred during the window.
The time (in microseconds since epoch) of the first order in this size range that occurred after this window, if no executions occurred during the window.
The transaction size, denominated in the currency of this order book, of the first order in this size range that occurred after this window, if no executions occurred during the window.
The trading system in use.
The trading mode in use.
The type of trading platform in use.
The best bid price available within the window.
The best offer price available within the window.
Table 4 is provided in a single file per day
The name of the order book referred to by this row.
The mean transaction price for transactions in this order book on this day.
The volume weighted average transaction price for transactions in this order book on this day.
The highest executed price for transactions in the order book on this day.
The lowest executed price for transactions in the order book on this day.
Supplementary information regarding costs can be found on the LMAX Exchange website, here and here.
Table 5 is provided in a single file per quarter.
Please note: LMAX Exchange does not offer rebates, discounts or other payments as part of its operational model, so this data can be assumed to be zero.
The name of the order book this row refers to.
The total costs of trading in the order book across this quarter, denominated in the currency of this order book.
The percentage of the total transaction value made up by the above costs, given to five decimal places.
Table 6 is provided in a single file per day.
The name of the order book this row refers to.
The number of maker orders accepted by this order book.
As above, but for market takers.
The number of trades that occurred on this order book.
The total value of the above trades, denominated in the currency of this order book.
The number of order cancellations made by makers on this order book.
As above, but for market takers.
The number of modifications to orders made by makers to this order book.
As above, but for market takers.
The median size of a trade on this order book, denominated in the currency of this order book.
The median order size of maker orders on this order book, denominated in the unit of the order book.
As above, but for market takers.
The number of market makers who participated in pricing this order book on this day.
Table 7 is provided in a single file per day.
The name of the order book referred to by this row.
The time this row refers to.
The best bid price at the specified time, or "-" if no price is available.
The best bid quantity at the specified time, denominated in the unit of the order_book.
The best ask price at the specified time, or "-" if no price is available.
The best offer quantity at the specified time, denominated in the unit of the order book.
The value of the available liquidity within three price increments of the current mid price, denominated in the currency of the order book.
Table 8 is provided in a single file per day.
The name of the order book this row refers to.
The mean of twice the difference between every transaction's price and the mid price at the time the order being filled arrived. "N/A" if no transactions in scope.
The mean volume, denominated in the unit of the order_book, of the available volume at the best bid.
A sample is taken at the end of each order book 'action'.
No sample is taken if no best bid or ask is present as separate metrics are presented to show average quote presence, etc.
"N/A" if no samples were taken.
As above, but for the volume at the best offer price.
The number of times a cancellation from a maker cancelled an order that, at the time of cancellation, was contributing to the volume and the best bid or offer price.
As above, but for taker cancellations.
The number of times a modification from a maker cancelled an order that, at the time of modification, was contributing to the volume and the best bid or offer price.
As above, but for taker modifications.
The mean value of the available liquidity within three price increments of the current mid price, denominated in the currency of the order book.
A sample is taken at the end of each order book 'action' where a two sided price is available at the end of that action.
"N/A" if no samples were taken.
For each transaction on the order book, the mean time between the aggressing order's arrival into the exchange and the time of the transaction. This will be "N/A" if no transactions occurred.
For each transaction on the order book, the mean time between the aggressing order's arrival into the exchange and the time of the transaction. This will be "N/A" if no transactions occurred.
It is thought that the above two columns negate the need to provide the "average speed of execution for unmodified passive orders at best bid and offer", as a limit order that matches the current best bid or offer will aggress (and fill) in the same way as a market order would; indeed, orders of that sort are included in the above two columns.
The number of fill or kill orders that failed to fully match.
The number of failed immediate or cancel orders that failed to match at all. Partial matches are not counted towards this total.
The number of trades valued at 100,000 EUR or more. Valuation to EUR is undertaken via rates derived from pricing data from LMAX Exchange.
The value of trades valued at 100,000 EUR or more, denominated in the currency of the instrument.
The number of trades values at less than 100,000 EUR.
The value of trades valued at less than 100,000 EUR, denominated in the currency of the instrument.
The number of interruptions to trading in the order book. LMAX Exchange prevents trading in certain market conditions automatically and may evict market makers thought to be providing incorrect prices. This can cause a short interruption to trading.
The average (mean) duration of interruptions, to the nearest second, or "N/A" if no interruptions occurred.
The number of suspensions.
The nature of the suspensions. This will always be 'manual', as suspension is administrated by the LMAX Exchange market operations team.
The average (mean) duration of suspension, or "N/A" if no suspensions occurred.
The number of periods longer than fifteen minutes where the order book was open but a two sided price was not available.
The average (mean) duration of periods without a two sided price that were longer than fifteen minutes, to the nearest minute. "N/A" if no such periods occurred.
The percentage of the order book's normal operating hours during which a two sided price was available. "N/A" if no operational hours occurred on this day for this order book.
We do not intend to publish Table 9 as it pertains to request for quote venues.