14.II.1 Head count

A main variable of equity pre- and post-trade data products – as apparent in the commercial terms between a data user (end-user or data vendor) and data supplier (direct feed or data vendor)1 – is the so-called head count. The term ‘head count’ refers to how the use and user type of equity pre- and post-trade data is discounted in the data price. To make this more specific, it is useful to look at the head count through five models of equity pre- and post-trade data prices. Five models to determine head count are: (1) per message/quote/ticker, (2) per user ID, (3) per device or terminal, (4) per application or (5) per business activity.2

The first model, being per message, quote, or ticker3 (model 1), indicates that the prices for the data are based on the volume of the data received.4 The last model, that is – per business activity, refers to the situation where data prices are based on the activity of the business, such as trading versus research (model 5).5 The other models, being per user ID (model 2), per device or terminal (model 3), and per application (model 4), are best understood against the background of how equity pre- and post-trade data is supplied to data users. As examined in part III of the research, equity pre- and post-trade data can be supplied directly from the original data source (e.g. an RM), for example, through the Internet (e.g. by means of so-called APIs).6 Equity pre- and post-trade data can also be supplied in an indirect manner, such as through data vendors. Very often end-users obtain the equity pre- and post-trade data indirectly via data vendors.7

The indirect supply of equity pre- and post-trade data results in several devices or terminals receiving data. For example, where an end-user of data simultaneously has access to two different data supplier terminals (two terminals) and to an API from one of both data vendors (one device), under a per device/terminal model, the end-user would be charged three times for the same data.8 The situation is different where a per user ID-model is used (model 2). In this situation, the end-user is charged only once rather than three times.9 Other examples of a head count are possible too, such as the situation where a per application-model is used (model 4).10 In this situation, the price for equity pre- and post-trade data is based on the amount of applications using the equity pre- and post-trade data.11 Examples include an application for a matching engine, an application for a trading algorithm, and so forth.12

Overview of data (information) flows in the EU equity market. Market participants can obtain the data directly from the platform, here referred to as ‘(Lit) trading venues’ (e.g. an RM). Investment firms (‘brokers’) and clients of the investment firms (‘investors’) can obtain the data directly. Other trading platforms, such as MTFs using reference prices from an RM, can also be a data user that obtains the data directly (here the MTF would be a ‘non-price-forming trading venue’ consuming equity pre-trade data and producing equity post-trade data after the trade on the MTF). Investment firms (brokers) and clients of investment firms (investors) can also obtain the data indirectly from data vendors, the latter compiling the data received from the source (e.g. RMs or MTFs).