Digging a Little Deeper

Galen Stops digs a little deeper into the results of the recent JP Morgan e-trading survey and finds some surprising statistics.

For those of you who missed it, there were some noteworthy nuggets of data contained within JP Morgan’s recent e-trends survey. But digging a little deeper beyond the headline figures reveals some even more interesting trends emerging in the FX market.

The first thing to point out is that the survey raises some curious questions about algo usage amongst clients. On the surface, it presents good news for algo providers – although only 8% of respondents said that they currently use algos for execution, 24% said that they plan to increase their usage of them in 2018.

Source: e-Trading Trends for 2018 survey, JP Morgan

These figures appear to chime with anecdotal evidence about buy side firms increasingly looking towards algorithmic execution to demonstrate best execution, help reduce market impact and take more risk at a time when some banks have pulled back to a degree from offering risk transfer services.

However, during the 2017 JP Morgan survey, 12% of respondents said that they use algos, with 38% saying that they intended to use more in 2017.

There is a clear discrepancy here, but what’s causing it?

Scott Wacker, head of e-commerce sales and marketing at JP Morgan, suggests that the 2018 statistics have been heavily affected by a change in the population of clients that JP Morgan surveyed. The 2017 survey was limited to just FX market participants, whereas this year it was expanded to include rates and commodities traders, which could have skewed the results in this area.

Indeed, Wacker actually reports seeing an increase in algo trading on the FX side of the business.

“We are actually seeing a shift towards the usage of algos, when and where it makes sense. Each client has a different need and desire to access liquidity in different ways. So an asset manager might use more passive strategies, hedge funds might use more aggressive strategies, banks might use algos to manage their order book, traders at banks will use internal algos to manage risks that they’re working themselves as a trader while they’re running risk. So the usage of algos is growing,” he says.

The other headline statistic that was widely picked up after the latest survey was released was that 61% of respondents said that they were at least “somewhat likely” to use a mobile trading app this year, up from 31% who said the same last year.

Source: e-Trading Trends for 2018 survey, JP Morgan

This is clearly a massive change, and perhaps even more significant is the jump in the number of respondents who said they were “extremely likely” to use a mobile trading app, from 18% in 2017 to 37% this year.

“I think what we’re seeing here is people getting more and more comfortable with mobile devices, with the security protocols, how it’s configured and the reliability of the platform. This is quite encouraging, and it does underscore a trend that is emerging. It’s still a relatively very small base of clients that are using mobile, but there is a growing appetite to accept mobiles devices for trading,” says Wacker.

The comfort level with mobile trading is clearly growing at a rapid rate. On the one hand, the ubiquity of smartphones in day-to-day life and the ease with which trades can be now be completed using these devices naturally leads to wider adoption. On the other hand, some of the more dramatic initial fears about mobile trading – for example, that someone would sit on a phone and make a trade by accident or decide to trade while drunk – seem to be subsiding as they have thus far not been realised.

Source: e-Trading Trends for 2018 survey, JP Morgan

More than trepidation amongst employees about using mobile devices for trading, it seems like company policy is likely to be the biggest barrier to mobile adoption in the coming years. But even this is evolving, according to Wacker.

“Company policy is concerned with how to control who is on the mobile platform and whether or not people should be allowed to trade while out of the office. Some firms still bar people from trading on mobile but allow them to monitor orders, read research and send in level suggest orders, but the level to which these restrictions are self-imposed is diminishing,” he says.

Another interesting quirk in the JP Morgan survey data centres around the electronic trading of EM currencies and NDF products. In the 2017 survey, 26% of respondents said that they trade EM currencies electronically and 32% said that they planned to trade more over the course of that year. But this growth in EM currencies does not seem to have materialised with only 27% of respondents saying that they currently trade them on electronic platforms in the latest survey.

Source: e-Trading Trends for 2018 survey, JP Morgan

Likewise, 24% respondents of last year’s survey said that they planned to trade more NDF products electronically, yet this year’s survey shows that only 9% of respondents are currently trading them electronically and the percentage planning to do so more in the future is 16%, suggesting that demand for e-trading this product type has decreased over the past year.

Source: e-Trading Trends for 2018 survey, JP Morgan

Again, the change in the type of firms responding to the survey might have distorted the results here, because Wacker is insistent that the volume of NDFs trading on JP Morgan’s electronic platform has increased significantly, particularly since January 3 of this year, when the Markets in Financial Instruments Directive (MiFID) II came into effect.

Because NDFs are in-scope of the MiFID II requirements, which are pushing certain products onto electronic platforms, Wacker says that JP Morgan has been comparing the NDF volume trends of the past few years to the first three weeks of 2018 and has seen business shift towards electronic trading.

“So for in-scope products, particularly ones on the vanilla side, we see a shift towards e-trading that I don’t think is quite reflected here. This could be more because the questions were viewed as being about whether there will be more activity in G10 versus EM. It could also be due to the change in the client base surveyed. But we are actually seeing a material jump in the ADV of NDFs, FX options and FX swaps – all of which are in-scope – on the electronic portals,” comments Wacker.

Source: e-Trading Trends for 2018 survey, JP Morgan

A final interesting data point from this year’s survey comes from the responses about how many and which type of electronic platforms JP Morgan clients are using.

A majority, 88% of respondents, said that they use a single-dealer platform (SDP), up from 71% last year, while 31% said that they only use an SDP, down from 36% in 2017. Meanwhile, institutional traders said they used, on average, 4.7 trading platforms over a 90-day period, an increase compared to an average of 4.4 platforms in last year’s survey.

Wacker says that this data makes sense, given the way that the market has evolved and matured. He claims that SDP activity has not decreased, but that with the multiplicity of different execution channels available – SDPs, multi-dealer platforms (MDPs) and direct APIs – clients are less likely to rely on one platform. Indeed, Wacker suggests that this is a good thing, stating that in many cases, clients are best served by not relying completely on one provider or channel.

However, another trend that he highlights is that JP Morgan clients tend to shift their trading activity back towards the banks’ SDPs when volume and volatility is on the rise.

Wacker explains: “The first is market impact – when you’re trading with one entity, your activity is less telegraphed out to the market. The reality is that these markets are already extremely transparent and depending on how you trade on a multi-dealerportal, your activity might be visible.

“The second is certainty of execution. When you’re trading with an entity on a single-dealer platform, you’re accessing the liquidity of the dealer and their entire franchise on a bilateral basis. A single dealer will prioritise this trade channel over all others so during active markets you’re more likely to get a trade done at the level you want.

But during periods of lower trading volatility, when not much is happening in the markets, clients can use a multi-dealer platform and there shouldn’t be too much difference in the outcome of their trades.”

Overall, the survey provides the basis for picking out a few important trends in the e-FX space. But given the change in the client base that the survey was sent to this year, it is hard to directly compare apples to apples. Hopefully the same format and client base will be retained for next year so that the seemingly surprising statistics around the responses regarding algo adoption, EMFX and NDF can more effectively be corroborated – or disregarded.

Galen Stops

Share This

Share on facebook
Facebook
Share on google
Google+
Share on twitter
Twitter
Share on linkedin
LinkedIn
Share on reddit
Reddit

Related Posts in