Siju Daniel,  Chief Commercial Officer - FXCM Group
Siju Daniel, Chief Commercial Officer - FXCM Group

Price isn't everything: separating good and bad liquidity in todays FX market

Providing pricing in the world’s most liquid market is a more competitive business than ever before – and brokers must ensure their customers have access to only the best quality liquidity from the most trusted counterparties, says Siju Daniel, Chief Commercial Officer - FXCM Group.

The diversity of today’s FX industry exacerbates this challenge. Post-crisis regulatory reform and appetite for risk warehousing has led to a steady decrease in banks’ market making activities, and non-bank firms have stepped in to fill their shoes. When combined with higher capital costs and reductions in leverage ratios, the resulting effect is a dramatic level of liquidity fragmentation across primary trading venues. How can brokers and institutional trading firms navigate this complex environment and guarantee good quality pricing from liquidity providers? And what key attributes must top-tier liquidity providers have in order to underpin and complement their client relationships? 

The answer to these questions is highly nuanced, but stepping back and surveying the broader market landscape reveals a number of critical metrics that must be considered when selecting a liquidity provider.

The changing face of liquidity provision

In the most basic terms, liquidity in the FX market pertains to a currency pair’s ability to be bought and sold without causing significant change in its exchange rate. A currency pair is said to have a high level of liquidity when it is easily bought or sold and there is a significant amount of trading activity for that pair.

Not all currency pairs are liquid. Currencies tend to have varying levels of liquidity depending on whether they are major, minor, exotic pairs or emerging market currencies. Typically speaking, liquidity dries up as a trader moves from major pairs to minor pairs and finally to the exotic pairs. Whatever the currency pair, choosing the right liquidity provider is a critical step in the execution process. 

Liquidity provision in the FX market has evolved over time with non-bank participants now accounting for a significant share of order entry and volume traded. As bank dealers, driven by balance sheet constraints, have pulled back from the market, these non-bank, non-traditional liquidity providers have emerged and proven extremely adept at filling the void created on the main trading venues. 

The recently released 2019 edition of the Bank for International Settlements’ (BIS) Triennial FX Survey paints a picture of healthy market, with a 29% jump in daily trading volumes to a record USD 6.6 trillion from USD 5.1 trillion in 2016. This growth has primarily been attributed to the rise of new proprietary and high-speed trading firms, an increase in FX swaps activity and more demand for emerging market currencies. Banks trading with “other financial institutions”, which BIS defines as non-reporting banks, hedge funds, proprietary trading firms, institutional investors and official sector financial institutions, grew significantly to USD 3.6 trillion, accounting for 55% of the global total. However, while this dramatic shift in market structure has had a positive impact on the growth of the industry as a whole, it has also introduced a number of complex challenges when it comes to accessing liquidity.

Navigating a fragmented market

In addition to a huge divergence in the type of institutions providing FX liquidity, the last five years have been marked by significant innovation in trading technology, resulting in the market fragmenting its execution activity across a growing number of venues. This has had a positive impact on the evolution of the industry, but in turn has made liquidity access and provision more complex than it was previously. 

FX trading is no longer restricted to the small selection of venues it was a decade or two ago. The market has moved from a model where all types of firms were forced to use the same execution protocol, whether it was appropriate for them or not, to a proliferation of innovative new platforms that meet the diverse needs of all participants.

A major driver of this diversification has been improvements in the tools available to measure market impact, in particular transaction cost analysis (TCA), meaning market participants and brokers are now better equipped to measure execution quality.

In addition, the growth in popularity of aggregators rather than single-venue screens means there is now less pressure on desktop space as traders can view quotes and orders across multiple trading venues without having to commit themselves to a single broker or platform. Aggregation, however, is not a catch-all solution to the fragmentation challenge. Surprisingly, trading on the best aggregated, on-screen price doesn’t always provide the optimum result. It is important that brokers aggregating pricing from multiple liquidity providers fully understand the quality and characteristics of the liquidity their customers consume.  Separating good and poor-quality liquidity is far more important than simply hitting what appears to be the best price.

FXCM
Liquidity provision in FX has evolved with non-bank participants now accounting for a significant share of order entry and volume traded

Price isn’t everything

So, if a trader can’t rely on price alone, what metrics should an FX broker consider when choosing which institutions are able to provide liquidity on their platform?
If the liquidity provider is a bank, one of the first factors to analyse is its ability to internalise flows. Many large banks in the FX market now regularly internalise client flows, often using their own electronic platforms. It has been estimated that as much as 30% of FX volume is internalised and internalisation can reach levels of over 90% for large dealers.

Large banks act as principals in these internal markets, and in some cases a bank’s internal liquidity book can be as big as the turnover of a major FX trading platform. As banks are able to quote tight bid-ask spreads, the liquidity conditions in an internal market can be overwhelmingly favourable. As a result, liquidity is often higher in the bank’s own internal market than in the external FX market. In addition, the price impact of FX trade execution in an internal market can be much smaller and less persistent than in the open FX market.

Analysing how much of an order a liquidity provider fills and what their pricing looks like post-execution can give an indication of a liquidity provider’s ability to internalise flows, and therefore whether the price a trader is seeing on an aggregation platform is good quality liquidity. If an order isn’t completely filled and the liquidity provider isn’t able to internalise that flow, they will start to unwind the amount they did fill and their pricing will reflect that. The market impact of the partial execution will increase with the ripple effect of the liquidity provider unwinding the flow. This is a hugely important metric to understand in order to assess whether the price provided is truly the best execution option.

There are a number of other very important factors that need to be taken into account. Brokers often fall into the trap of thinking that onboarding more and more providers leads to tighter spreads. While that may be true in the short term, it is not a strategy that will make for long term, consistent, high quality execution. Typically spreads will either eventually widen, or execution quality will deteriorate.
 
On one hand, brokers must closely monitor their providers to ensure that they are meeting the necessary requirements, but on the other hand they must work hand-in-hand to help them do the best job that they can. Doing so will allow them to provide tighter spreads and a higher execution quality for customers. 

A large part of this task is sending a profile of flow that they expect to receive (and are pricing for), as well as grouping them into a pool alongside other providers that use a similar hedging model. Brokers must be very careful when deciding who can execute with the providers, and who is executing alongside them in the same pool, as both can ultimately end up affecting the quality of execution.
In order to maintain a select list of providers, brokers must not only ensure that their liquidity providers’ execution is of high quality, but also consider their coverage of instruments, coverage of time and whether they will provide reasonable execution during highly volatile markets – when some providers tend to stop pricing.

Some of the metrics that should be monitored include:

  • Fill ratio/rejection ratio: This is simply the rate at which the provider fills or rejects. It is helpful to monitor this rate both as a percentage of volume as well as orders. A provider may do well executing small orders but reject any order of larger size, so their rejection rate between volume and orders could vary significantly.
  • Round-trip time: This is the time that it takes a provider to respond to an order request. Longer hold times generally result in higher negative slippage for the customer. It is also helpful to split this measure between fills and rejects. In many cases there is a very low reject rate. This means that performance on rejects can get drowned out when looking at overall numbers. In order to maintain a high quality of execution, reject metrics generally need to be split out and focused on separately.
  • Cost of rejects: This measures the movement of the market after an order has been rejected and estimates the price impact on the customer. This can be correlated to the round-trip time of rejects. A higher cost of rejects would generally lead to higher slippage.
  • Wide spreads: The performance of a provider during volatility is very important, and knowing whether a provider will offer reasonable liquidity during high volatility is critical. Some providers may widen to an absurd spread in order to maintain a high-time coverage metric, but are essentially invalidating and removing themselves from executing.
  • Quote count: A provider sending excessive quote updates can cause several issues. This can create performance issues, but also the ability to hit the actual price being shown. If the provider sends a new update in the time it takes the order to be received, then they may consider that the broker is routing on an old quote, rather than them rejecting a legitimate order. In addition, the quote count specifically during news or high volatility events is important to isolate. The market is quiet more often than not, which means terrible performance during a news event that lasts a few minutes may not be picked up when looking at the aggregate stats for the day.Market share: This is also an extremely common metric, and often well correlated with spread. But it can also provide more information than spread alone. Let’s say a provider quotes a tight spread, but when volume picks up they widen out. Their ranking in terms of general average spread may be very different to their ranking in terms of market share.
  • Top of book: While also very similar to spread, this can be a little more telling. Let’s say a provider has a comparatively wide spread, but skews significantly. Then they could still be at the top of the book on one side, tightening the aggregated spread. The time spent at the top of the book on either side may be more telling of a provider’s impact on the customer’s spread than the provider’s spread itself.
  • Time coverage: In addition to the general coverage through the day, a broker must check whether a provider covers times like the market open and the minutes around rollover. These are typically very low-volume periods that providers either can’t or prefer not to cover. But from a customer standpoint, the broker should provide a smooth consistent experience by providing reasonable spreads and liquidity during these periods.
  • Instrument coverage: Maintaining a smaller list of important providers means that the coverage provided by each is also a consideration.
  • Market impact: Measuring market impact shows which providers have similar hedging models and would work well together in the same pool. There are customers and providers in many different categories, so matching is very important. Mixing them will result in the entire pool ending up with the metrics of the worst performing category.

TCA: the foundation for success

Effective TCA is a critical tool in today’s FX markets to measure quality and pricing of execution against the metrics outlined above. TCA provides concrete data that brokers can use to assess their liquidity providers and allows them to set a consistent framework that ensures like-for-like comparisons. This information can not only inform a broker’s liquidity provider selection process, but also encourage providers to adjust their pricing in order to remain competitive. The ability to gain deep insights into liquidity provider performance and improve the customer trading experience can make or break a broker in today’s challenging FX market conditions. Just as electronic trading heralded an entirely new era for the FX markets almost three decades ago, sophisticated data analysis tools and expertise will have a similar impact over the next few years. Understanding this is the critical foundation for a successful modern FX brokerage business.