The Effects of Time Zones On Cross-Border Liquidity
Harrison Mann
,
Head of Growth
FX is a 24-hour market.
That’s true in a technical sense. At any given time, a currency can be traded somewhere in the world. This characteristic makes FX distinctive: Unlike equities, there's no bell-toll for the closing auction, no moment where everyone heads home for the night.
But “always open” and “always liquid” aren’t remotely close to the same thing.
This “24-hour market” is a cascade of time-distinct overlaps as major trading centers like Tokyo, London and New York go on and offline. During the overlaps, particularly the London-New York window that runs from 1pm to 5pm London time, liquidity is deepest. Spreads are tight and orders generally fill cleanly.
Outside them, the picture rapidly changes. Spreads widen and order books go thin, a trade that would have executed smoothly at 2pm London time might gap the market at 2am.
As we know, the global economy doesn't run on London time. Payments need to move when businesses operate, and for a major portion of the world's most important payment corridors, the times when liquidity is thinnest are exactly when the need to transact is highest.
Who was the market built for, anyway?
Large institutions in major financial centers mostly, trading predominantly G7 currencies. The deep liquidity that characterizes EUR/USD, GBP/USD, USD/JPY and their relatives exists because of decades of infrastructure investment: Electronic trading platforms, clearing systems, prime brokerage relationships, and market-making desks. So the market that arose was one that excels at serving those participants in those currencies, at the hours that suited them best.
The “Asian session,” roughly Tokyo and Singapore hours, delivers deep liquidity in JPY, AUD, NZD and the major Asian currency pairs. The “European session” is where EUR, GBP and CHF come online. New York picks up USD flows and keeps things liquid through the afternoon. As mentioned, during the London-New York overlap, almost everything trades well.
Things are much less ideal at the margins. Late New York into early Tokyo is the thinnest window of the trading day, the stretch from roughly 5 to 7pm New York time, where neither hemisphere is fully online. Even in major pairs, spreads noticeably widen. In so-called “exotic” pairs, the effect can be severe.
Consider where the world's highest-volume remittance corridors actually sit. Within the US-Mexico corridor. one of the largest in the world by volume, the Mexican peso trades well enough during US hours, but thins out beyond them. The US-Philippines corridor involves the Philippine peso, which is predominantly traded during Asian hours.
West African corridors involve currencies whose primary liquidity windows don't neatly align with either European or American business hours. The same is true across much of South and Southeast Asia, along with Latin America outside Mexico.
These corridors aren't marginal. They represent hundreds of billions of dollars in annual flows. That’s millions of people sending money home, businesses paying suppliers, and companies managing cross-border operations. The infrastructure we have built does not work for them.
The true cost of thin hours
The practical consequences of time zone liquidity gaps are extensions of the problems we’ve seen elsewhere when liquidity begins to dry up.
The most direct cost is wider spreads. Market makers price liquidity according to risk, and thin hours mean higher risk. When there are fewer participants to absorb an order, and less ability to hedge quickly, market participants pay in the form of higher transaction costs.
For a remittance business processing thousands of transactions a day, a spread that's 20 basis points wider during off-peak hours is a significant cost that then gets multiplied across the entire book.
The second cost is execution uncertainty. In thin markets, the price you see and the price you get diverge more frequently. Slippage increases, and orders that would fill instantly during the London session might take seconds or minutes during the Asian period, by which point the price has moved. Once again, these costs are passed onto the party making the transaction.
The third cost is more structural: Forced timing. Businesses that understand liquidity gaps will batch transactions around peak windows, so their payment operations end up getting driven partly by market microstructure as opposed to business logic. A company in Lagos shouldn't have to think about the London-New York overlap when deciding when to pay a supplier. But if the spread on their currency is worse outside that window, the market is essentially imposing a tacit schedule onto them.
This is the inequity embedded in time zone liquidity gaps.
A business in London or New York trading EUR/USD doesn’t have to think this way. Their currency pair is liquid around the clock, or sufficiently so that the gaps are inconsequential. The further you move from that default, the less “24/7” the FX market starts to feel.
A problem, we believe, is worth solving.
FAQ
If a currency can be traded somewhere in the world at any hour, what does it actually mean to talk about a trading "session"?
A "session" is just the stretch of the day when a given region's financial centers are at their desks. The trading in each window is dominated by the banks and market makers physically working in that region during their own business day. So the Asian or European "session" refers to a concentration of activity that follows office hours around the globe, not a venue that opens and shuts. The market is continuous only because these regional working days overlap at their edges and hand off to one another.
Why is it cheaper and easier to exchange a currency at some times of day than at others?
What you pay depends on how many market makers are actively quoting at that moment, and that number rises and falls with the clock. When two major regions are at their desks at once, the largest number of firms are competing to quote at the same time, so more parties are willing to take the other side of your trade and the cost of transacting falls. When only one region is working, fewer firms are quoting, each is carrying more of the risk. The price responds accordingly. The cost of moving a currency is, in effect, a measure of how many people happen to be awake and trading it.
Why can one currency be easy to trade at an hour when another is nearly impossible?
A currency's liquidity is tied to the part of the world where its economy and its banks operate, because the institutions most willing to make a market in it are the local ones, trading through their own business day. The Mexican peso trades well during North American hours and thins out afterward, the Philippine peso is deepest during Asian hours. A currency is most tradeable when its own region is the one awake to carry its flows.
When is the worst time of day to trade?
The thinnest stretch is the seam between the New York afternoon winding down and the Tokyo morning ramping up, roughly 5 to 7pm New York time, when neither hemisphere's financial centers are fully staffed. With the fewest market makers active, even heavily traded pairs show visibly worse pricing, and minor or exotic ones can become hard to move at any reasonable cost. The quality of a trade tends to depend less on the size of the order than on how many participants are at their desks when it lands, and that window is the low point of the day. It's also when an unexpected headline can move prices furthest, since there are fewer participants around to absorb the reaction.
Does the FX market ever actually close?
Yes, the "always open" description applies to the business week, not the calendar. Trading runs continuously from roughly Sunday evening, when the Asia-Pacific centers come online, through Friday evening in New York, with the regional working days handing off around the globe and no nightly break in between. Over the weekend, with financial centers everywhere shut, there's effectively no trading and no live price. That's why a gap can open between where a currency closed on Friday and where it reopens on Sunday: whatever happened in the meantime gets absorbed all at once on the reopen.
Why would a business end up scheduling its payments around financial-market hours rather than its own needs?
Because the cost of converting a currency can differ dramatically between peak and off-peak hours. A company moving money in a currency with a narrow liquid window will often cluster its conversions into that window, batching payments to hit the hours when pricing is best, rather than transacting whenever the business would naturally prefer. In this way, market structure ends up setting the payment calendar for firms across the globe. For companies dealing in G7 currencies this problem barely matters, since those pairs stay liquid enough around the clock, but the further a currency sits from that default, the more the clock starts running the schedule.
Share article
Read other articles
Stay informed with our latest articles on currency launches, institutional FX trends, and global liquidity.
