About a-team Marketing Services

A-Team Insight Blogs

Breaking the Dominance of the 4pm Fix: Alternative Approaches to FX Benchmarks

Subscribe to our newsletter


Since the mid-1990’s, foreign exchange (FX) benchmarks have played a crucial role in providing standardised FX valuation points, which are used for a wide variety of purposes by financial markets participants across the industry. Over the last decade however, particularly in the wake of the Libor scandal of 2012 and the FX fixing scandal of 2013 – both of which involved benchmark manipulation – some of the most widely-used industry benchmarks have come under intense scrutiny.

This has led to the emergence of new regulatory standards, new technologies, new calculation methodologies and even new benchmark providers, all aiming to improve the accuracy, transparency and integrity of benchmarks and how they are used.

So how are the key FX benchmark providers responding to this changing landscape, and what does the future hold for these critical financial instruments?


FX benchmarks were originally created to provide a common valuation point, particularly for fund managers, explains Colin Gallagher, Multi-Asset Index Product Manager at Bloomberg Index Services Limited.

“Decades ago, it was recognized that when portfolio managers were buying and selling foreign bonds and equites, they created foreign exchange exposure. They were facing imbalances in valuations across a portfolio as the transactions were being done at different rates during different times of the day,” he says. “The concept of one common rate to value the portfolio across, was born. For a decade these rates were referred to as ‘fixings’, but nowadays they are labelled as financial benchmarks with administrators and regulators as actors in the space.”

The first tradable FX benchmark, launched in 1994, was the Reuters WM/R Benchmark. Rebranded as the Refinitiv WMR FX Benchmark in 2018, it remains the most widely used FX benchmark in the industry.

“The original intention around WMR is still very important, even now,” says Shirley Barrow, Global Head of Refinitiv Benchmarks at Refinitiv, an LSEG company. “It originated to provide a trusted source of FX rates that were reflective of the market, from an independent provider, to provide a level of transparency that wasn’t there before. Since then, the use cases around WMR have grown and expanded. It’s now used in all kinds of situations where a client needs a standardised, point in time, FX rate that’s reflective of the market.”

The 4pm fix

The majority of FX benchmark trading occurs at the UK ‘4pm fix’, which became dominant because it aligns with the closing of the equity markets and provides fund managers with a common valuation point for fund accounting, performance measurement, and portfolio rebalancing.

“If you’re an equity fund manager for example, and all you care about is equities, and you want to have zero tracking exposure, you have to trade at the 4pm fix,” says Jamie Walton, co-Founder of FX benchmark provider SirenFX, a relatively new player in the benchmark space. “Plenty of funds who are aware of some of the problems associated with the fix have to track it, as part of the fund’s mandate,” he adds.

The problems that Walton refers to mainly stem from the concentration risk of so much trading happening in such a short amount of time. It’s not unusual, for example, for the entire day’s trading range to occur in the run up to the fix, with the high or low of the day occurring at 4pm.

“One of our goals is to increase choice,” says Walton. “Having one dominant benchmark causes a concentration effect, because everyone is trading in that tight window. So we looked at what could be better for the end user, and set about creating a benchmark that doesn’t have that detrimental market impact or lead to concentration risk.”

Alternative Approaches to Benchmark Calculation

Walton describes how the calculation methodology of the Siren FX benchmark differs from more established benchmarks such as Refinitiv’s WMR.

“We use a different type of calculation, which is similar to that used in equities,” he explains. “If you want to trade ‘at the close’ in equities, you can’t trade at the close itself, so what people actually do is trade into the close, ramping up their trading as the close approaches. That’s quite a well-defined, mathematically sound ‘optimal execution’ method, based upon a mathematical model called Almgren-Chriss. For the calculation of the Siren benchmark, we start off with a low weighting for the spot at the start of the 20-minute window and then we increase the weighting as we get towards the end. The key thing,” claims Walton, “is that this gives you a more accurate benchmark with less market impact than something based around TWAP (time-weighted average price), for example.”

In order to address evolving market requirements and to ensure that their WMR benchmark continues to meet users’ needs, Refinitiv last year conducted a Request for Feedback exercise, where respondents were asked questions focusing on three key areas: the length of the benchmark calculation window (currently five minutes, increased from one minute in 2015); the data sources used for the benchmark calculation; and ideas for new benchmarks. Shirley Barrow discusses some of the findings.

“Regarding the calculation window, some thought five minutes was fine, others suggested that it could be a little bit wider, particularly perhaps at peak times such as month end, which is a very busy time,” she says. “It’s a balance between the potential for concentration risk if it’s five minutes versus the potential for increased volatility and tracking errors if it’s a longer window, and we have to get that balance right. We also need to make sure that any change we make is very thoughtful and considerate of our users. We’ve certainly been spending a significant amount of time evaluating and looking at different scenarios as to what the impact would be of a wider window.”

Walton adds his thoughts on Refinitiv’s Request for Feedback. “It’s notable that a significant number of the respondents thought that there should maybe be a longer window,” he says. “So there is clearly demand out there for alternative approaches to publishing a benchmark. That was very interesting, and points to a potential future where those clients could be using things like Siren as an alternative. The request for feedback was good in the sense that it took the temperature of where the market is, and the feedback will help LSEG/Refinitiv – as well as ourselves – produce something ultimately better for the client.”

Data sourcing

In order to calculate the WMR benchmark for what it refers to as the 25 ‘trade currencies’[1], Refinitiv sources transaction and order book data from its own FX trading platform, Refinitiv Matching, as well as from EBS and Currenex. “With regard to data sources, one of the things that we want to make sure of is that we are sampling the market as broadly and as representatively as we can,” says Barrow. “Refinitiv Matching and EBS are certainly significant right now, but we see new platforms demonstrating a growing share of liquidity, so we’ve started some dialogue in that space and are doing some initial assessment on what currencies are available and what depth of liquidity they have on these platforms.”

Bloomberg’s BFIX benchmark is based on trades and orders that take place on the widely used Bloomberg Terminal and is derived from data from multiple liquidity providers and trading platforms.

The Siren FX benchmark uses data from New Change FX, an FCA-authorised benchmark administrator that also sources its pricing data from multiple FX liquidity venues, providing continuous, officially regulated benchmark rates for Spot and Forward FX.

“Our aim is to be transparent, so for example you can see the underlying rates we use from New Change FX, what the benchmark calculation currently is, and you can measure any tracking error, all in real time,” says Walton. “You can also see almost immediately what the final benchmark is. It’s all public. There’s nothing secret.”

Addressing benchmark manipulation

The Libor and FX fixing scandals, which came to light in 2012 and 2013 respectively and resulted in billions of dollars’ worth of fines for the offending banks, placed benchmark manipulation in the spotlight. So what steps are regulators and benchmark administrators taking to prevent future manipulation and to ensure the accuracy and fairness of benchmarks going forward?

“The FSB report on FX benchmark reforms in Oct 2015 was an impactful release to the market as they stated, ‘the report re-emphasises that the FSB recommendations are intended to apply to all FX benchmarks, not just the WMR 4pm London fix’,” says Bloomberg’s Colin Gallagher. “They emphasized the word all by putting it in italics, and also made note that central banks should at least take note of guidance from the IOSCO principles for financial benchmarks.”

He continues: “These changes have led to today’s safer environment where index administrators have considerable expertise managing and producing FX benchmarks under governance oversight and production controls. Banks have ring-fenced the process of fixing orders, and compliance and market surveillance teams are comfortable they have controls and checks in place.”

Barrow explains Refinitiv’s approach to assuring the integrity of the WMR benchmark. “The WMR benchmark plays a fundamental and important part in decisions, in valuations, and in products, so it has to be maintained at a very credible standard,” she says. “As such, we look at things in a multi-layered way to promote that integrity and make it difficult to manipulate. Our operations team runs 24 hours, five and a half days, and we have a robust quality validation process that runs constantly, validating the accuracy of the data. We also have a completely separate and segregated monitoring and surveillance team, and we run a dedicated third-party application, which produces alerts focused on behaviours in the market and provides a full escalation process.”

Looking Ahead

FX benchmarks clearly play an important role in the financial markets. As such, in order to remain fit for purpose, they need to continue to evolve with the market. So what might that evolution look like?

“Automation of spot FX has been going on for decades but other FX products, namely FX forwards, have struggled somewhat to achieve the same efficiencies,” says Gallagher. “It is this area that requires investment to be able to offer asset managers the ability to hedge their projected cashflows and reduce tracking error across their portfolios for those. Every month when vast sums of Spot FX flows are generated from portfolio rebalancing there are also huge FX forward flows to hedge but they are not necessarily accepted as orders against FX forwards benchmarks due to most of these flows not landing on standard tenors, which requires interpolation. As interest rates have started to move in the past year this issue has been amplified and many FX benchmark providers are seeking to provide solutions to overcome this.”

He continues: “As more and more emerging market central banks become more familiar with financial benchmark best practices there will be additional scope to move to independent third-party benchmark providers offering opportunities around the globe to expand. With these opportunities on the horizon, there is a healthy outlook for FX benchmarks.” Colin Gallagher, Bloomberg

The industry certainly seems receptive to new FX benchmark providers, if somewhat slow to embrace them. “Often the driving force is the asset owner, rather than the asset manager,” says Walton. “It’s the pension funds seeing that they could maybe do better on their FX hedging if they were to use an alternative benchmark. And it’s a relatively easy change to make, but there is a little bit of inertia in the market, people are often happy with the status quo, but this may be because they are not aware how much this inertia is costing them.”

The last word goes to Shirley Barrow. “What’s really important is that we’re not complacent, we need to ensure that our benchmarks remain fit for purpose and evolve, and user engagement is obviously a core part of that as well. I hope and I certainly intend to continue providing FX benchmarks to the quality that users need.”


Subscribe to our newsletter

Related content


Recorded Webinar: Best practice approaches to trade surveillance for market abuse

Breaches of market abuse regulation can lead to reputational damage, eye-watering fines and, ultimately, custodial sentences of up to 10 years. Internally, market abuse triggers scrutiny of traders and trading behaviours; externally it can undermine confidence in markets and cause financial instability. This webinar will discuss market abuse of different types, such as insider trading...


Why Automation is Going to Dominate in the Race to T+1

By Vikas Srivastava, Chief Revenue Officer, Integral. Tick-tock tick-tock, the clock is ticking down to one of the most pivotal moments in financial markets this decade. This coming May will mark the transition to T+1 settlement for equities and bonds transactions in North America. To add to the complexities, CLS confirmed last week it will...


AI in Capital Markets Summit London

The AI in Capital Markets Summit will explore current and emerging trends in AI, the potential of Generative AI and LLMs and how AI can be applied for efficiencies and business value across a number of use cases, in the front and back office of financial institutions. The agenda will explore the risks and challenges of adopting AI and the foundational technologies and data management capabilities that underpin successful deployment.


GDPR Handbook

The May 25, 2018 compliance deadline of General Data Protection Regulation (GDPR) is approaching fast, requiring financial institutions to understand what personal data they hold, why they process it, and whether it is shared with other organisations. In line with individuals’ rights under the regulation, they must also provide access to individuals’ personal data and...