LP Relationship Management
LP Management Best Practises
For all types of Trading participants, STP, A/B, Market makers, good LP relationships are an important factor in running a profitable Trading business.
What you should look for in an LP?
An LP can be judged by their trading characteristics:
Spread
Markout / market impact
Interest to fill / cancel
Last look period
Fill Ratio
Continuity of rate provision (do they gap when market events happen?)
Asymmetry of their last look (do the markouts of their rejects materially differ from their fills)
Matching process e.g. FastMatch
Anonymous or Disclosed Venues
Execution costs
Order size capability
Full Amount or Sweep streams
Simplicity of NOP and margin requirements
This in turn can vary depending on a number of factors including:
Instrument / Asset class
Timezone e.g. singapore hours vs london hours
News events
Unexpected news events and market volatility
Market state
LP Number
With too few LPs the aggregated spread is wider. As you add more LPs the varying price skews they supply can reduce the aggregated spread. Too many LPs negative selection occurs or “winner’s curse” where they only receive the flow when their rate is vulnerable. In this case, LPs in turn end up rejecting the flow, resulting in having to replace the order with the dollar cost of rejects driving up the effective spread of the pool.
LP Curation
Curating the optimal set of LPs for your price formation can be challenging and Mahi‘s suite of technology can assist with that.
Analytics
To make LP relationships work you have to have analytical capabilities for tracking the above characteristics and see how they vary by changing conditions.
Automation
Another key element is automation.. Tooling to uncover the sophisticated ways your connections to LPs can be abused, which when this happens you will see it in your future spreads that you receive and show clients.
Alerting
Checking this behaviour should be automated to reduce the time from identifying a negative change in trading SLAs from your LPs to being able to action that.
Period on period comparison
Not every change in trading SLA may be sufficiently big to generate an alert but it is important to swiftly identify changes in behaviour and track them week to week / month to month.
Regular liquidity reviews
Weekly or monthly reviews of your LPs should be conducted to ensure that you are creating a healthy set of liquidity.
It's a relationship
If you are receiving undesirable liquidity from an LP, then it is important to check-in with that LP as to why the undesirable behaviour is being conducted on their behalf and if there’s anything you can do to collectively result in better liquidity being offered in the future.
Where concessions are being made on your behalf it may be worth asking for better service in other areas.
User tagging
Sometimes LPs ask you to enable user tagging where the originator of the flow is disclosed to them. Sometimes it can be useful to enable this to ensure access to better trading SLAs.
MahiFX
MFXEcho helps you understand LPs via its comprehensive analytics and is able to check for changes in trading SLAs from your LPs.
Compass is a pricing, risk management and hedging solution. It allows its users to offer clients different pricing and trading SLAs than those received from LPs allowing user’s to take control of their own trading SLAs and not relinquish that to the LPs.
It offers varying degrees of execution quality in its hedging strategies allowing users to control how much they want to exploit vulnerabilities in an LPs pricing.
Where risk-transfer trades have been executed at a favourable rate for the LP, it also offers the capability of reclaiming that PnL using proprietary trading signals. A sustainable model that can ensure that the LP still has some profit and incentive to continue to provide good trading SLAs and benefits yourself by bounding risk at better value than may have arisen due to the timing of onboarding that risk.
Opaque Markets
Retail books can see more flow than banks. It's one of the harder parts of making in FX. Its quite an opaque market. Monthly volumes of $2.8B yards at Exness alone. At scale, retail sentiment is correlated. If you find that you have lots of risk in one direction, it is likely that other brokers have that too.
=> Risk transfer trades as you hit position limits. Are not necessarily non-informational i.e. $25M dropped to an LP can be the exact kind of flow they fear. Possibly you want to opportunity hedge your risk ahead of those limits
Questions for your LPs
Tell them which aspects of their Trading SLAs are uncompetitive and what improvements they can make in order to ensure a greater percentage of the trading flow is routed to their business.
If they are going to change the spread distributions that you receive can they give you notice and opportunity to take remediative steps.
What is their last look period?
Do they utilise symmetrical or asymmetrical last look?
Do they pre-hedge during the last look window?
Do they utilise rejected order information as an input into pricing?
Questions that an LP provider may ask of yourself?
Q) Can you enable user tagging so we can ask for selected counterparties to be removed from our liquidity?
An element of judgement should be taken into account.
Where flow originates from Compass hedging strategies, it is possible to give this flow with no tag. Alternatively, it is possible to:
supply a single tag
rotate across a range of tags with which we can change periodically
use a random tag
attribute the tag of the last client to trigger the risk