Once again, it’s that time of year when our editorial staff dust off the infamous Profit & Loss crystal ball in order to take a peek into the future and tell our readers what they should expect from the year ahead.
Colin Lambert’s “Trade of the Year” makes a welcome return, and he’s back with a bang as he focuses on the drivers of the ever-popular NOK/MXN pair. As has become custom, Lambert is also predicting consolidation within the FX industry, but regular readers might just be surprised to find out that for once he doesn’t think that the M&A activity will be on the platform side this year.
Meanwhile, Galen Stops stubbornly insists that CTAs will turn around their recent run of bad performance and come good in 2019 and offers a foreshadowing of what’s to come in the upcoming Bank for International Settlement’s upcoming triennial OTC FX volume survey. Whether or not readers agree with these predictions, they offer some thoughts on the key talking points in the FX industry right now.
1. More Algos from a New Source – Colin Lambert
In his first prediction for 2019, Managing Editor Colin Lambert believes we will see more algo execution products rolled out into the market, but not from the incumbent players.
If there is one challenge the algo has to overcome in 2019 it is negative publicity. Just about every time something happens in FX markets the headlines are rolled out about algos being out of control, or triggering flash events. It is, of course, hard to really nail down whether they are either the former or indeed play a role in the latter, but the headlines do help to feed a fear of the unknown amongst some customers, thus making them more reluctant to explore the benefits of algo execution.
The trouble for these clients is, however, without a pretty radical change in policy at the regulators, then they may well have to get used to algos, because, well, they’ll have little choice. The seemingly unstoppable growth in demand for best execution and for everything to be measured means that ultimately the business of risk transfer, or working an order for a client becomes simply uneconomic for most banks. Of course they would love to continue to execute in this fashion, but the demand for best execution, allied to rules around capital allocation, mean the numbers don’t add up.
In these circumstances, the easiest way to retain customers (beyond the fact they hold the credit purse strings) is to roll out your own suite of algo execution tools – and that is what I think, finally, a large swathe of regional banks are going to do. Some have already done this, of course, SEB and Commerzbank both spring to mind as banks with these product sets available to customers, but 2019 will be the year in which many more jump on the bandwagon.
After all, if you can’t make money by trading with your customer, then become an agent and earn a fee – until of course competitive pressures come to bear. It may not be the preferred route for many regional banks, but it is likely to be the only reasonable way they can retain the vast majority of their FX customer base.
The key differentiator for regional banks is local know how and inventory, and for this reason the algos they roll out have to be more attuned towards accessing internal pools of liquidity rather than the more prevalent agency model, which hits any and every venue on the street. This means there will not be a whole host of new algos with rather catchy (or corny) names, there will be a TWAP, a Peg, an Iceberg, and, perhaps, a VWAP (they probably will have silly names however).
The impact of these new algos will be interesting because previously the algo world has been the domain of the global giants in FX banking circles, and now they will find their own volume levels dropping because a huge piece of their current client base – regional banks – will be using their own products (or rather their customers will). Quite how this plays out vis-à-vis the relationship between regional and global bank is unclear, but given how both sides have become more sensitive to criticism over the past year or two, it will not be totally positive.
This could play into the hands of the public venues as the new generation of local algos seek liquidity, however I don’t think it will have that big an impact – and the reason is simple.
Just as their new algos will have to access their internal liquidity pools in their specialist pairs, the relationship model can be lifted from the broader FX business. Regional banks do not pump all of their local market liquidity down the global banks’ pipelines, they reserve them for currency pairs in which they are not experts, so their suite of algos will be all about the specialist currency pairs. Liquidity is vitally precious in the modern FX market and I do not believe that regional banks – reliant as they are upon their LPs in other pairs –will jeopardise those relationships by having their clients use their algos for something like GBP/USD. Besides, in an interesting reversal, we may even see some global banks using the regional banks’ algo suites to execute risk in certain currency pairs.
Whether or not these new algos will be a success will not be known for a few years, but there is only one way to find out. Global banks are becoming more aggressive in their marketing of their algo suites and when a treasurer or execution desk head has to produce a TCA report to “prove” best execution, it is so much easier when you are using algos that use exactly the same data. Yes, there is a credit issue there occasionally, but more often than not it is actually just about the relationship – the global bank has not yet got to regional centres with its sales force yet.
This means the local banks need to catch up, if only to where the global banks were two or three years ago, for when 2020 dawns, if a regional bank does not have the most basic of algo execution strategies available to their clients, they will find that client base thinning out.
2. Traditional Financial Services Will Buy Into the Crypto Space – Galen Stops
Predicting M&A activity amongst FX platforms is an easy one to make, which is perhaps why our managing editor, Colin Lambert inevitably makes it each year. Too easy, I say. Instead I’m going to make the case that we will see at least one firm from “traditional” finance buy one that has emerged within the crypto space.
Despite the collapse in the price of cryptocurrencies over the past year, I remain bullish about the long-term prospects of this market. Personally, I’m less concerned with price – I think a pretty solid case could be made that $3,500 (the price at the time of writing) is still a crazy valuation for one bitcoin – and more focused on what can be achieved through tokenised assets and building blockchainbased solutions on the networks underpinning cryptocurrencies.
In principle, I even like the now muchmaligned ICO model, which if regulated properly could enable more firms to secure funding while giving individuals an opportunity to make meaningful investments in early stage firms, rather than getting the leftover opportunities once the venture capitalists and professional investors have had their fill.
I think that enough firms in traditional finance see these opportunities as well and are already committed enough to developing crypto offerings or services of some sort that they won’t be deterred by the price action. At least one firm from traditional financial services will opt to buy a crypto platform, wallet provider, market maker, etc, for any combination of these reasons: (a) they lack the requisite expertise in this market (b) to augment their own internal projects in this area (c) they want to wait and see which services and solutions the market coalesces around and how regulations shake out (d) it’s just easier for them than building in the crypto space.
3. A Stall, Not a Reversal – Colin Lambert
A popular theory in markets is that as banks pull back from some of their traditional services such as market making, the buy side will step in. Colin Lambert believes quite strongly this won’t happen.
There has been much discussion in fixed income markets about big asset managers becoming market makers in certain very liquid bond markets and more than a few people have expressed the same thought in the FX market – that the buy side will step up and fill in liquidity gaps that have appeared in recent times.
I do fully expect some firms to dip their toe in the water, and I have no doubt that many customers will take advantage of their banks’ willingness to open their order books to them so they can place bids and offers in the market (thus providing liquidity indirectly), but actually become market makers? Not a chance.
There is a fundamental flaw in the argument that buy side firms can become market makers and that is in the raison d’etre of those firms. They are investors or manufacturers and FX remains a service to them, not an asset. Investors are not interested in punting FX markets, they want to hedge their very deeply considered investments in bond and equity markets, and in those circumstances the last thing they want is to be placing random bids and offers in EUR/USD. How does an asset manager explain that they underperformed their peer group because they got smashed on a price in USD/JPY?
In fact, given the obsession with tracking error, how does an asset manager explain how they outperformed because making markets in FX is so easy and lucrative (and yes, when it comes to Galen’s review of this next year, I am being sarcastic!)
The premise of this argument sits in the definition of market maker. This should be a participant who is willing to make a price in most reasonable currency pairs (let’s be generous and say just the G7) on demand. They cannot pick and choose when they put prices in (although some “liquidity providers” seem to do this) they must be there at all times during the major market hours.
This is anathema to a firm that has spent the last few decades as the client of a bank, which meant they were able to ask for a price at any time, without consequence. So expect to hear more chatter about the buy side stepping up to the plate as market makers, but as to whether it will actually happen? Not a chance.
4. FXPB Consolidation to Reverse (Slightly), Asset Manager Clients to Remain Elusive – Galen Stops
“Igo to all these industry events and there’s always a discussion around credit and when I hear them I sometimes I wonder how long we can keep stretching out this particular piece of gum, but the reality is that it’s still one of the biggest issues in the market today,” one market source opined in a recent conversation and in some ways I couldn’t agree more.
Following the SNB’s decision to remove its peg to the euro, which caught a lot of PB clients unawares and caused sizable losses, plus a broader reallocation of costs across banks, credit has undergone a significant re-pricing in recent years. One consequence of this was that a lot of FXPB activity consolidated into a small number of players, and I think it’s fair to say that Citi was probably the biggest beast in the jungle when this had all shaken out.
However, in December of last year Citi relocated its FX prime brokerage (FXPB) business from within its global FX franchise to its prime finance and securities services (PFSS) division, following losses that were valued at as much as $180 million related to loans to an Asian-based asset manager. While this needn’t necessarily precipitate a pull-back from Citi in the FXPB space, I still think that the consolidation of business that has occured in the past few years will start to reverse slightly in 2019 as firms with traditionally smaller PB businesses offset their lack of scale with a combination of efficient technology and reduced costs.
One thing that won’t change is that FXPBs will continue to struggle in their attempts to sell services to the asset managers and, more broadly, the real money community. FXPBs have been pitching to these clients based on regulatory changes that are forcing them to post margin with multiple counterparties, arguing that it would be simpler and easier to just amalgamate all of these businesses through into one PB relationship.
However, speaking to some FXPB salespeople at the start and end of 2018 there didn’t seem to be much progress on this front. This client segment is a notoriously long sell-cycle and, because there’s already been clearing requirements for certain products, many of their operations teams have developed efficient collateral management solutions, including for margining. Neither of these things will change in 2019.
5. If You Can’t Beat ‘Em… – Colin Lambert
The rise of the non-bank market maker has been well-documented and 2019, suggests Colin Lambert, could the year that this manifests itself in the M&A market.
There is, if enough study is conducted, a well-worn path to a takeover deal. Two firms compete in a market, they then find they have specific advantages they can exploit, but that does not do enough for them to win the battle, they then start to think about working together. This may come in the form of joint ventures in certain geographies or markets, it may come with shared support facilities. The last stop on this journey is merger or takeover.
Take this theory into financial markets and have the competitors being the banks and the non-bank market makers. They have most surely competed and continue to do so, and they have specific advantages in balance sheet strength and technology, respectively, but, as they have inevitably found out in recent years, they need each other still.
We have, especially in fixed income and equity markets (and to a lesser degree in FX) had specific collaborations, for example between Virtu Financial and both JP Morgan and Bank of New York Mellon. What then, the chances of a more formal link up?
At face value the chances don’t seem that great – after all, the non-bank firms have succeeded to an extent that even some of their founders find mildly surprising. They have been helped by postGFC regulatory changes and the reluctance of a growing group of banks to actually allocate risk to their financial markets businesses, but succeed they have – and that makes them expensive acquisitions. Surely it must be easier to build your won if you are a bank?
Actually it isn’t because of the sheer size of the infrastructure required and the resources required, particularly financial. It is one thing allocating money for an M&A deal that the Board can trumpet as progress, quite another coming up with the money to fund a technology project to make the firm a fraction of a second quicker at pricing and hedging.
This means that where before they were not open to such an idea, more banks are likely to at least informally discuss the pros and cons of buying a non-bank market making firm. The resources needed to maintain the high levels expected in a nonbank firm in terms of technology can be maintained currently, so a bank, with larger resources should be able to fund it very easily. The key will probably be maintaining the non-bank unit as just that – a separate unit. Many are the innovative businesses that have withered and died once placed inside the bureaucratic infrastructure of a behemoth bank.
So a pet “non-bank” market maker operating under the balance sheet (but little else) of a bank is no longer the distant prospect it was – after all they are profitable businesses, with relatively low cost bases and remain scalable. More pertinently, they are, in the main, simply better at making markets and hedging (smaller) risk than banks are.
For the potential targets there could also be a change of mood, although it is hard to believe that the founders and owners of some of the non-bank firms would not bite the hand off if an offer came along at any time in their existence. Whereas before it just wasn’t considered because the founders either didn’t want to work in a big institution (again) or were willing to be patient because their inevitable success would command a bigger cheque from a buyer, times and circumstances have changed.
It is hard to tell, apart from the size of balance sheet, between the top end nonbank firms and the medium sized banks. Their respective regulatory burdens (relative to size of the firm) are converging and the non-banks have sucessfully moved into more asset classes, which will get progressively harder, so now is perhaps the time they see to bank that nice big cheque.
The starting point may not be a mega deal between a bank (and it should not be ruled out that a large asset manager thinks of acquiring a non-bank player) and one of the top echelon players, but there are plenty of other firms with the technology resources and existing access to markets that may suddenly become attractive targets.
A sensible deal would appear to involve one of the top echelon players, however there may be the question of the price tag to contend with there. As always in the banking industry they like someone else to take the dive before joining in, and that may mean a smaller deal before a bigger.
One other factor that may drive a deal is the changing market structure in non-FX markets. Credit is becoming less of an issue in many asset classes and that is levelling the playing field for the non-bank firms. This democratisation of the markets will only likely continue and that means these firms become even more valuable – meaning the time to buy is now, not in two years’ time when they are even more successful.
6. CTAs to Stage a Comeback – Galen Stops
Ok, so this prediction is made more in hope than anything else. In the Q3 edition of the magazine last year Colin and I debated whether or not trend following, a strategy utilised by many CTAs, has had its time as a strategy, with him claiming that it doesn’t work anymore and me taking the other side of the argument. CTAs endured a pretty miserable 2018 and funds utilising trend following strategies in particular underperformed, with the Societe Generale Trend Index finishing the year down 9%.
However, regular listeners of our In the FICC of It podcast will know that when Colin challenged me regarding our debate that I doubled down, claiming that my argument was right, but that he was looking at too short of a time horizon. Hence, I’m predicting that CTAs are going to have a good 2019.
There are in fact some positive signs out there, not least the US stock market which after a seemingly endless rise is finally having some wobbles. A combination of continued trade wars, the end of the tax-cut induced economic sugar high and the likelihood of even more political dysfunction in Washington means that the US equities markets will struggle to replicate the performance of the 2009-2018 period. And CTAs historically tend to perform well during equities drawdowns because as firms sell equities, it causes other asset classes to sell-off as people look to buy safety assets.
So these drawdowns create volatility expansion, creating trends and movement out of consolidation areas as investors rush to raise or lighten exposures. Plus, CTAs can jump on the trend of shorting the equity markets. This, plus the fact that they always seem to perform just when everyone is writing them off, is my basis for predicting that 2019 will (finally!) be a good year for CTAs.
7. A New Definition – Colin Lambert
The year 2019 started off in fine style for FX markets, with what was widely termed a “flash event”, but Colin Lambert argues that by the end of the year such moves will be more commonplace and not be seen as “flash”.
A 400 point fall in a currency pair in the FX market is not a new phenomenon – it has happened throughout the history of the foreign exchange market, thanks to events mainly. The last decade has seen the invention of the phrase “flash event” to describe these moves, mainly because, as it is in life generally, things happen much quicker than they used to.
The thing is though, are these moves really any different to what used to happen? The answer is no, beyond the speed at which they take place. It could actually be argued that the advent of data analytics and algos means that the FX market actually sees less flash event than it could perhaps if it were mired in the old structure.
Executions are (generally – ignoring the chance that some idiot will try to do an “amount” in the Australasian window) smarter now and take into account likely market impact. Without this quality of data traders and machines just five or 10 years ago were largely executing into the dark. Once every three years there was a measure of liquidity in FX markets thanks to the BIS survey, a decade ago this was reinforced by the semi-annual surveys from the world’s major FX committees, but even these admirable attempts to provide data of the depth of the market fail to deliver because they are on an aggregated, daily, basis. Who knows how much USD/JPY can be sold at 3pm New York? Well nowadays a number of decent analytical tools can give you a good idea (although the oldest adage about liquidity – you can get as much done as you like when you’re wrong – still stands).
There is also the fact that more firms are aware of these moves and are quicker to spot them. Without a real driver in terms of actual news, there will not be another 14 big figure, or 30 standard deviation event in a major currency pair, not least because there are now dozens of traders eager to jump on what will be the inevitable mean reversal when a scan of the newswires returns nothing.
The FX market still has to come to terms with the fact that it is operating in a more unstable era geopolitically and macro-economically – an environment probably last seen in the 1990s, a decade that brought us European and Asia monetary crises and the junk bond debacle (actually very late 80s that one) and the dotcom bubble. In other words, markets are going to move because uncertainty is at very high levels compared to recent history when the relatively tranquil nature of markets (as long as you weren’t involved in credit) helped boost the profile of automated trading and risk management.
In such an environment we will see bigger, and fundamentally justified, moves in prices and these moves will come quicker. Financial markets do not just randomly move on their own – someone has to buy and sell to gets things going and that will happen more and more frequently in 2019. In this scenario, there will be larger moves, that happen quickly, and are not totally reverted – it is significant that the move lower in AUD/USD for example at the start of 2019 was only about 60% reverted, mainly because the net move was an appropriate response to the sudden “risk off” mentality in markets.
It will be impossible to stop the headline writers and social media sensationalists from referring to these moves as “flash” crashes or rallies, but amongst more professional circles they will be seen for what they actually are – a considered reaction to a fundamental change in conditions.
As much as some politicians don’t like it, there is no stopping markets moving, for as long as conditions change they will react accordingly. What the world has to get used to now is the actual speed at which this information can be consumed and reflected in pricing levels. If a “risk off” scenario existed in the pre-automated trading days that meant, for example, AUD/USD would fall 200 points the market would indeed have fallen that far, probably over a 30 minute time horizon. The market would also, probably have overshot and dropped another 100 points over a five or 10 minute spell, before coming back.
In modern markets the reaction would be the same, it’s just the time horizon – there may be a 300 point move in three minutes and within another two it would have reverted. Same rationale, same price level, same result – different pace.
So December 31, 2019 will be the cause for reflection in FX markets and it will be seen as a more volatile year in terms of price action, but also one in which the market handled just about everything thrown at it. Importantly, though, moves will not be instantly referred to as “flash”, more we may go back to the future and commentators will say, “markets were busy”.
8. FX Will Get Back to Growth in the BIS Survey – Galen Stops
The 2016 Bank for International Settlements (BIS) central bank survey showed that average daily FX turnover was $5.1 trillion, down 5.5% compared to 2013. This was the first time that trading volumes had contracted since the 2001 survey showed a 19.5% decrease from the 1998 one, but I expect the 2019 survey to show that this was a temporary dip rather than a decline.
Now I know what you’re thinking – just saying that the BIS number is going to be higher than 2016 is way too easy, especially when we’ve been looking at the semi-annual central bank surveys which have already suggested this will be the case. So to make things harder, let’s try and put some numbers on this.
I’m going to say that the headline volume figure will be between $5.8-6.1 trillion, a 13.8-19.6% increase.
Why? Because, despite the concerns about the growth of populist political rhetoric and the increasing advocacy and implementation in some quarters of protectionist and isolationist policies, global trade – still the fundamental driver of the foreign exchange market – continues to grow. Granted, in the back half of last year the World Trade Organisation (WTO) revised down its forecast for global trade growth in 2018 to 3.9%, citing rising trade tensions and the potential monetary policy tightening and associated financial volatility to destabilise trade and output as the key reason for this, but it’s still significantly higher than the 1.3% growth recorded in 2016.
On top of this, divergent monetary policies have opened up more opportunities for firms looking to make a profit in the currency markets, while there will be enough geopolitical factors to keep markets moving throughout the year. If nothing else, for those who are able to trade it effectively, expect Brexit to be the gift that keeps on giving.
A few other predictions. In the 2016 survey hedge fund and prop trading activity had declined significantly to account for just 8% of total FX trading activity and while life hasn’t gotten any better for hedge funds the prop trading firms will have become a larger part of the overall market so I’m going to say this segment will be in the region of 12%. Last time round swaps trading grew 6% and this figure will more than double in the 2019 survey, FX options trading will get back to growth after a 24% decline in volumes in the 2016 survey and RMB trading will continue growing at a very fast rate but slower than the 81% registered between 2013 and 2016.
9. A Proper Trade of the Year – Colin Lambert
Last year’s Crystal Ball saw, for the first time, the absence of a trade of the year from Colin Lambert – something he wants to put right this year.
For those readers who have scanned P&L editor Galen Stops’ and I’s critiques of each other’s performance you would have picked up on a theme in mine about how much Galen sat on the fence with his calls last year. One such prediction was his call on EUR/USD, which vexed me greatly, not only because he was stepping into my domain, but because he hid behind expert analysis. More to the point it was EUR/USD! Where’s the fun in that?
2019 is therefore, I feel, the year in which tradition must be restored and as someone who has gone on the record in one of our In the FICC of It podcasts as saying I like my hedge funds a little crazy I feel it beholden upon me to put on a trade that accurately reflects the proud traditions of this annual event – it must be random, bizarre and in a currency pair that is only seen when correlated hedging mechanisms go horribly wrong.
Welcome then readers to a “proper” trade for 2019 – buy NOK/MXN! At the time of writing the cross is about 2.22, so we’ll call it 2.2150 for a quick start and we are buying this because…hang on, I’ll think of something…oil prices will tighten and risk aversion may heighten. That sounds like a good reason to buy Nokkie and sell Mex – frankly that’s about as scientific as it will ever get – although for those of you now rushing out to put the trade on I do have a word of warning and it involves correlations.
Over all the years I have been making trade predictions my record is balanced, in that I have got as many right as wrong (I may actually be net up on percentages thanks to a couple of good trades involving the Brazilian real and Kiwi dollar). Those that I have got wrong, often to the tune of at least 10% all have one thing in common – they involved that FX trader’s Kryptonite, a Scandinavian currency.
In all my 40-plus years in the foreign exchange industry I have never understood how people make money trading these currencies, but obviously they do (local knowledge is a real advantage and on the basis I travel to the region at least once every two years I feel qualified), so I am jumping back in the water. The sharks may be waiting and they may be hungry, but if nothing else it’s going to be more fun tracking this than some boring G5 pair.
10. Full Amount Trading Will Increase in 2019 – Galen Stops
When pressed about whether liquidity conditions in the FX market are better or worse than they used to be, a fairly common assessment that I’ve heard is that there’s more liquidity than ever in small size at the top of book, but that it’s becoming increasingly hard to get large amounts done.
Combined with a growing concern about market impact when trading, this has led to the increased use of algos for execution, in order to chop up trades into smaller pieces to find good liquidity and avoid moving the market. Now the logical conclusion of this trend would seem to be that as more people slice up their orders then it exacerbates the original trend that caused them to do this, meaning that orders are getting increasingly small but more frequent.
Thus, while it seems inevitable that algo trading will continue growing in prevalence as the tools become more sophisticated on the one hand and as firms become more comfortable using them on the other, I think that 2019 could see many firms lean further back towards full amount trading.
The first reason is that, perhaps contrary to popular belief, improved transaction cost analysis (TCA) tools that are becoming more readily available to a wider set of market participants will show that in many cases clients will actually receive a better price by trading the full amount they need with one counterparty and opposed to sweeping a range of platforms and trading with numerous different counterparties. Rather than bank vs. non-bank, liquidity providers are increasingly being evaluated through the prism of risk taking vs. nonrisk taking, and clients will find that trading full amount with risk taking liquidity providers is the best way to reduce their noise in the market. And for their part, liquidity providers will in some cases prefer to trade on a disclosed basis to clients that they know want a real price from them and where they know exactly how much they’re looking for.