About over-the-counter markets
As you know, the Forex market is an over-the-counter market (the so-called OTC market). This means that, unlike exchange markets with a single center, where each transaction has two parties – the client who sells the asset and the client who buys this asset, in OTC markets the Dealing Center is always the second party to the transaction. The dealing center accumulates thousands of multidirectional orders from its clients and, on its own behalf, re-places transactions in electronic trading systems, operating at the same time with its total position. Therefore, it is often impossible for each specific deal of a dealing center’s client to match a specific deal on the counterparty’s side. Clients’ transactions are placed in the electronic trading system (interbank market) in batches, the so-called snaps, and may contain orders from several clients at once.
Straight Through Processing (STP)
Currently, the market uses two business schemes for building a brokerage company in the OTC markets – STP and a hybrid model. Let’s consider them in more detail
It should be noted that now, in pure form, these models are practically not used separately. Both methods of working with client positions are used in parallel. The distribution of positions and their actual withdrawal to the interbank bank are usually regulated automatically using the so-called liquidity bridge, special software that connects the dealing center server and liquidity providers.
With the STP model, all client transactions are automatically relayed by the dealing center to various trading systems. You can also find a synonym for the STP model, NDD service (Non Dealing Desk). In other words, when using the NDD service, each client’s transaction, regardless of the volume, market environment and financial result of this transaction, is renegotiated on another counterparty. The counterparty is usually liquidity aggregators such as Curenex or Intergral. But it can also be just a larger forex broker or a separate bank.
It should be noted that all transactions, both in the case of the NDD and in the case of the hybrid model, are, in accordance with established practice, exhibited anonymously. That is, not on behalf of the client, but on behalf of the dealing center. Moreover, orders are often placed by the dealing center on the interbank market in batches, so-called “snaps” and include orders from several clients at once. That is, the liquidity provider cannot determine which particular client has placed a given trade. Thus, price manipulation on the part of the liquidity provider and setting the “worst” prices to a specific client of the dealing center is completely excluded.
Such anonymous placing of transactions became possible due to the almost widespread use of the so-called FIX protocol, according to which the dealing center in an encrypted anonymous form and exposes the transactions of its clients to electronic trading systems.
The advantage of this scheme is that absolutely all risks are completely transferred to the liquidity aggregator, the total financial capacity of which is much greater than the capacity of the dealing center. Consequently, the reliability of this system is close to 100% (if the liquidity providers are selected correctly).
The disadvantages of this business model are that the dealing center is almost completely excluded from the process. A dealing center can only earn in one way – by putting its own interest in the price. Therefore, almost every deal will be closed worse than the market, sometimes quite significantly. Also, the disadvantage of the pure STP model is that it is impossible to guarantee accurate execution of limit orders at the stated price and uncontrolled spread expansion during high market volatility (although on average the spreads with this model are lower than when using the classic hybrid model).
In addition, under the STP model, it is difficult to offer a service for clients with a small deposit size, since liquidity providers have their own requirements for a minimum transaction. With a large deposit and large order volumes, there may also be problems with execution. The waiting time for the execution of the transaction increases, because it is not always possible to find a counterparty for a large volume at a favorable price. Also, with this model, it is difficult to implement various trading options and marketing programs, since dealing is assigned a passive role.
That is why, at the moment, dealing centers usually use together STP and the classic hybrid-netting model.
Hybrid netting model
With this business model, the dealing center operates not with each client deal separately, but with the aggregate position of the company. At any moment of time, not one client deal is opened inside the dealing center, but a multitude of multidirectional positions.
The superposition of all client deals for each trading instrument shows the resulting, final position of the dealing center. For example, if the broker has 500 contracts to buy euros and 300 to sell at the same time, the resulting position will be 200 buy contracts. Instead of making 800 trades, the dealing center can only withdraw the resulting position to an external counterparty at a time or in parts. This principle is called netting. And the model is a hybrid netting model.
The advantage of this model is that the company’s resources are significantly saved. Clients can get excellent trading conditions, quick conclusion of transactions, and the limitation on the minimum deposit is almost completely removed, since the company aggregates the entire mass of client orders and operates only on the aggregate position.
The total client position can be split into several parts and displayed to different counterparties, depending on the prices and volumes they offer at the current time. Elements of spatial arbitration are applied, that is, to search for the best counterparty during the delay in order execution. Aggregate large orders separately from small ones that completely overlap themselves, etc.
There are many techniques for working with the aggregate client position to achieve optimal trading conditions. As a rule, this task is automated within the framework of the Liquidity Bridge and is constantly being improved by correcting various weights and algorithms for processing statistical data. In each company, these tasks are solved slightly differently and are usually the broker’s know-how, although the common features described in this article are repeated for everyone.
Competent use of the hybrid model, in conjunction with NDD technologies (especially work on the FIX protocol), guarantees high reliability of the company while creating unprecedentedly comfortable conditions for any arbitrarily small or large volumes served by a dealing center.