The aggressive order flow from retail customers has a significant value. Inbound buy orders trade at the best ask price, while aggressive sell orders trade at the best bid price.

If a counterparty trades against the two orders by executing a day trade, its result will be the quantity traded multiplied by the size of the contract and by the difference between the ask and the bid price, i.e. the spread.

In several international markets there are rules that allow intermediaries to act systematically as counterparties to the respective retail customers’ order flows, providing liquidity to those flows.

The intermediary is the main party responsible for origination of the retail customers’ order flow. The greater its capacity for providing liquidity for this flow, the greater its incentives will be for promoting and developing the customer base and the markets’ liquidity.

In Brazil, Cross Trades have been used for some time in the WIN (mini Ibovespa Futures Contract) and WDO (mini US Dollar Futures Contract) market for brokerage house proprietary accounts to operate against the aggressive orders from retail investors.

Products with this feature have had a superior performance in recent years:

  • Greater customer base growth;
  • Greater ADV growth;
  • Greater liquidity at the best price level of the order book;
  • Reduced spreads.

This way, B3 developed a new order type that makes possible for every intermediary to be the counterparty to its retail clients’ order flow, providing more liquidity for WIN (mini Ibovespa Futures Contract) and WDO (mini USDollar Futures Contract) markets.

RLP features:

  1. Only intermediaries who meet the requirements established by B3 (transparency with customers, opt in and opt out mechanisms etc.) would be able to use RLPs;
  1. RLPs could be aggressed only by orders from customers of the same intermediary who were flagged as retail customers;
  1. RLPs would be market pegged orders (the intermediary would indicate the buy and/or sell quantity and the order price would be automatically adjusted by PUMA to the best bid or best ask);
  1. If the spread between the best ask and the best bid were two or more tick sizes, the RLP price would improve by one tick size or more at the intermediary’s discretion;
  1. Considering the top price level of the order book at a given time and the arrival of an aggressing order from an intermediary’s retail customer:
    1. The RLP would be ranked ahead of all orders from all other intermediaries (by price/broker/time priority instead of price/time priority). In this case the aggressing order would be forwarded to the RLP book. Any remainder left after execution in the RLP book would be sent to the central order book.
    2. The RLP would not be ranked ahead of orders from customers of the same intermediary that matched the aggressing order (there is no preemption for customers of an intermediary).

Thus, if an order from another customer of the same intermediary was aggressed on the opposite side of the book, the aggressing order would be routed to the order book and would be matched with existing orders from all brokerage houses up to the last order from the same intermediary’s customer. Any remainder after execution in the RLP book would be sent to the central order book.


  1. The aggregate volume of RLPs in the market would not be allowed to exceed Y% of the total volume of the instrument (e.g. 15%). Calculation of “total volume of the instrument” and “aggregate volume of RLPs” would exclude contracts traded by participants on their own account via RLPs and include contracts traded by customers of participants via RLPs. Given the above rule and considering that RLPs would only be for retail customers, each brokerage house would be able to submit at most Y/X of its retail volume as RLPs, where X is the market share of retail customers in the instrument considering the whole market.

Percentage Y could be changed by B3 depending on market conditions, with participants being notified in advance. Percentage X would be updated every month and calculated as the daily average market share of retail customers in the previous month. Thus, the average computed for month T would be the parameter to be observed as the limit for month T+1. The purpose of this limit would be to assure that a significant proportion of the retail flow originated by brokerage houses interacted with other investors, contributing to an increase in liquidity and a reduction in spreads.

  1. Because no orders in the overall market would be able to aggress them, RLPs would not have pre-trade transparency but would be disclosed via the market data feed immediately after trading.
  1. Through the BM&FBOVESPA PUMA Trading System, intermediaries A1, A2, …, An could allow intermediary B to act as counterparty for the purpose of “the respective flows of aggressing retail orders in compliance with all other conditions. Brokerage houses A1, A2, …, An could stipulate a fee to be paid by intermediary B.
  1. Through the BM&FBOVESPA PUMA Trading System, an intermediary could allow a customer to act as counterparty to its flow of aggressing retail orders in compliance with all other conditions. The intermediary could stipulate a fee to be paid by the customer.
  1. Initially, RLPs would be used only for Mini US Dollar Futures (WDO) and Mini Ibovespa Futures (WIN). In future, B3 would review the list of authorized products based on its assessment of the results of the new functionality.

Benefits of the RLP:

Best execution, latency and holding back orders

  • Elimination of the initial latency of the brokerage house’s OMS (necessary in facilitation);
  • Guarantee of obtaining the best price available on the market or better (pegged price or better)
  • Risk mitigation when deliberately holding back orders (simplicity of the OMS)

Trading Ahead

  • Assurance that trading ahead is not taking place over brokerage house’s customers orders (PUMA Trading System control). Assessment of the chronological sequence of orders exclusively based on the PUMA Trading System clock.


  • Easier BSM supervision of the brokerage houses
  • Less room for undue exploitation of conflicts of interest by the brokerage houses.

Democratization of the product and the effects of competition

  • Possibility of using the product by all brokerage houses with a retail flow without having to make major investments in IT.
  • Possibility of small brokerage houses selling their retail flows to “flow consolidating” market makers (elimination of the essential facility problem).

Price formation process

  • A cap to guarantee efficient operation of the price formation process.
  • Spill over to the book (cap and protection against trading ahead) generating increased liquidity at the top of the book and reduced spreads.

Investor protection

  • Assurance of best execution (absence of additional latency at the same price or better).
  • Authorization for use of RLP conditional upon information transparency and opt out mechanism.