#2341: Why the Dollar Rules Every Currency Trade

Why does trading Thai Baht against South African Rand rely on the dollar? Dive into the mechanics of cross-pairs and global FX markets.

0:000:00
Episode Details
Episode ID
MWP-2499
Published
Duration
24:36
Audio
Direct link
Pipeline
V5
TTS Engine
chatterbox-regular
Script Writing Agent
Claude Sonnet 4.6

AI-Generated Content: This podcast is created using AI personas. Please verify any important information independently.

How the Dollar Shapes Every Currency Trade

Currency trading might seem like a straightforward exchange between two currencies, but in reality, the dollar is almost always involved — even when it’s not part of the trade. This phenomenon is most evident in cross-pairs, where currencies like the Thai Baht (THB) and South African Rand (ZAR) are traded directly. Despite their bilateral economic ties, the exchange rate between these currencies is determined not by their direct trades but by their respective relationships with the dollar.

Cross-Pairs and the Dollar’s Role

A cross-pair is an exchange rate between two currencies that doesn’t involve the dollar directly. However, the rate is calculated by dividing the dollar exchange rates of the two currencies. For example, to find the THB/ZAR rate, traders divide the USD/THB rate by the USD/ZAR rate. This method dominates even when direct trades between THB and ZAR occur.

The reason lies in liquidity. The dollar is involved in 88% of all global FX trades, creating deep, liquid markets for dollar pairs. When trading minor or exotic currencies, the dollar pairs (USD/THB and USD/ZAR) offer tighter spreads and better hedging options than direct trades. As a result, banks and market makers prefer to hedge their exposure through the dollar, reinforcing the dollar’s central role.

Arbitrage and the Tether Effect

Direct trades between minor currencies are possible, but they operate in the shadow of the cross-rate. Arbitrageurs ensure that any deviation between the direct rate and the calculated cross-rate is corrected instantly. High-frequency trading algorithms monitor these discrepancies and execute trades within milliseconds, pulling the direct rate back into alignment with the cross-rate. This arbitrage force acts as a tether, preventing direct trades from setting independent prices.

The Cost of Illiquidity

Trading exotic cross-pairs comes with a premium. Spreads on pairs like THB/ZAR can be 10 to 20 times wider than those on dollar pairs, reflecting the cost of managing illiquid inventory. During periods of market stress, these spreads can widen dramatically, making direct trades even more expensive.

When Direct Markets Matter

There are exceptions. Regional trade corridors with dense bilateral flows — like THB/MYR (Thai Baht/Malaysian Ringgit) — can develop independent liquidity. Factors like local banking infrastructure, regulatory frameworks, and central bank swap lines also play a role. However, even in these cases, the dollar’s influence remains significant.

The dollar’s dominance in FX markets is a product of network effects. Its liquidity attracts more participants, which in turn reinforces its liquidity. While alternatives like the euro have emerged, the dollar’s role as the universal intermediary in cross-pair pricing remains unchallenged.

Downloads

Episode Audio

Download the full episode as an MP3 file

Download MP3
Transcript (TXT)

Plain text transcript file

Transcript (PDF)

Formatted PDF with styling

#2341: Why the Dollar Rules Every Currency Trade

Corn
Daniel sent us this one — and it's a question that's been sitting in the back of my head since he mentioned it. You're trading the Thai Baht against the South African Rand. Two minor currencies, real economies, real trade flows, and yes, some actual direct trading between the two. So why is the rate you see on your screen not really a Baht-Rand rate at all? Why does it live or die based on what the dollar is doing against Bangkok and Johannesburg separately?
Herman
This is where it gets genuinely strange, because most people's intuition is — if there's a market, the market sets the price. Direct trades happen, supply and demand, done. But that's not quite how it works in FX, and the reason why tells you something fundamental about how global currency markets are actually structured.
Corn
By the way, today's episode is powered by Claude Sonnet four point six — our friendly AI down the road handling the script.
Herman
Okay, so — cross pairs. Let's get into it.
Herman
A cross pair, at its most basic, is an exchange rate between two currencies that doesn't involve the dollar on either side. But the way it's calculated — and this is the part that matters — is by taking two dollar pairs and doing the division. If you know how many dollars one Thai Baht buys, and you know how many dollars one South African Rand buys, you can derive the Baht-Rand rate without ever executing a Baht-Rand trade.
Corn
Which sounds almost too simple. Like, of course you can do the arithmetic.
Herman
Right, but the question Daniel's really asking is — why does the arithmetic win? If there are actual THB/ZAR trades happening, why don't those trades set the rate?
Corn
Before we get there — what even makes a pair "minor" or "exotic" in the first place? Because those terms get used loosely.
Herman
The rough taxonomy is this. Majors are the seven or eight pairs that include the dollar against the euro, the pound, the yen, the Swiss franc, the Canadian dollar, the Australian dollar, the New Zealand dollar. Minors are pairs between those same major currencies but without the dollar — EUR/GBP, AUD/JPY, that kind of thing. Exotics are everything else — one major currency against a currency from an emerging or smaller economy. THB/ZAR is actually exotic-on-exotic, which is about as far from the liquid center as you can get.
Corn
You've got two currencies that are each already thinly traded against the dollar, and now you're asking what they're worth against each other.
Herman
And the market's answer is — we're going to calculate that, not discover it. The USD/THB rate and the USD/ZAR rate exist with real depth, real volume. You divide one by the other and you get THB/ZAR. The direct market exists, but it's working in the shadow of that calculation the whole time.
Corn
Why doesn't the direct market just...
Herman
Because it can't. The numbers don't let it. Here's the core issue — as of the most recent BIS Triennial Survey, the dollar is on one side of roughly eighty-eight percent of all FX trades globally. And about seventy-five percent of all trades involve a dollar pair as at least one leg. That's not a coincidence, it's infrastructure. The dollar is the hub and everything else is a spoke.
Corn
When you try to trade THB against ZAR directly, you're operating in a market that's essentially a rounding error relative to the dollar pairs on either side.
Herman
And thinness has consequences. The moment any direct THB/ZAR trade prints at a price that deviates meaningfully from the calculated cross-rate, arbitrageurs close that gap in milliseconds. High-frequency traders are running these calculations continuously — if the direct rate drifts even a fraction of a percent from USD/THB divided by USD/ZAR, there's a risk-free profit sitting there, and it gets taken immediately.
Corn
The direct market can't really go its own way because the moment it tries, someone yanks it back.
Herman
The arbitrage force is essentially a tether. It's not that direct trades don't happen or don't matter — they do — but they cannot sustain a price that's inconsistent with the cross-rate. The cross-rate is the gravity well.
Corn
Can you give a sense of the scale of that? Like, how fast are we actually talking when an arb opportunity opens up?
Herman
Sub-second, in most cases. The latency arms race in FX means that the major electronic venues — EBS, Reuters Matching, the big bank platforms — are running these triangular arbitrage checks essentially in real time. A deviation of even two or three basis points in a cross-rate will attract automated flow within a fraction of a second. By the time a human trader has noticed the discrepancy and thought about acting on it, the gap is already closed. The only people who reliably capture those moments are the ones who co-locate their servers physically next to the exchange infrastructure and have written algorithms specifically to hunt for them.
Corn
It's not like there's a window where a savvy trader can sit there and manually arb the cross against the dollar pairs.
Herman
That window existed in the nineties when pricing was still done over the phone and information moved slowly. Now the tether is essentially instantaneous. Which reinforces the point — the cross-rate isn't just theoretically anchored to the dollar pairs, it's mechanically anchored, in real time, by automated systems that never sleep.
Corn
What about from the market maker side? Because banks are in the middle of most of these trades, and they have choices about how they hedge their exposure.
Herman
This is where the economics get really clear. If I'm a bank and I've just bought a bunch of Thai Baht from a client who needed to sell — I now have Baht exposure I need to hedge. I can try to find a counterparty who wants to buy Baht against Rand, or I can just hedge via USD/THB, which has deep liquid markets, tight spreads, and I can execute large size without moving the price. The USD route is almost always cheaper.
Corn
Holding inventory in an illiquid pair is a cost the bank doesn't want to bear.
Herman
And that cost shows up in the spread you pay as a client. When a bank does quote you a direct THB/ZAR price, they're embedding a premium for the inconvenience of managing that exotic exposure. The cross-rate calculation gives them a floor — they know what the fair value is — and then they charge you extra for the illiquidity on top.
Corn
How much extra are we talking? Like, what does that spread difference actually look like in practice?
Herman
It varies a lot by pair and by market conditions, but to give you a rough sense — on a major pair like EUR/USD, you might be looking at a spread of half a pip or less in normal conditions. On something like USD/ZAR, maybe three to five pips. On a direct exotic cross like THB/ZAR, you could easily be looking at spreads that are ten to twenty times wider than the constituent dollar pairs. And that's on a good day. When there's stress in either underlying market, those spreads can gap dramatically — you might go from a ten-pip spread to a fifty-pip spread in minutes, because the market maker is now carrying inventory they really don't want and they're pricing that discomfort into the quote.
Corn
The spread on a direct exotic quote is basically the bank's fee for doing the USD plumbing on your behalf.
Herman
The third piece of this — which I think gets underappreciated — is that most pricing engines at major banks and electronic venues don't even start with direct price discovery for minor pairs. They calculate the cross first, then check if any direct market activity warrants an adjustment. The direct trades inform the spread and maybe nudge the mid-price slightly, but the anchor is always the cross.
Corn
NOK/SEK is a good example of this, isn't it? Because those two currencies have real direct trading — Scandinavia, integrated economies, actual cross-border flows.
Herman
Right, and it's instructive precisely because it's not exotic. Norwegian Krone, Swedish Krona, neighbouring countries, substantial trade relationship. You'd expect that pair to develop independent pricing. And there is real direct liquidity — more than THB/ZAR by a long stretch. But even there, the EUR cross remains the primary anchor. EUR/NOK and EUR/SEK are both liquid enough that the EUR-derived rate sets the tone, and direct NOK/SEK trading happens in its shadow.
Corn
Even when the conditions are about as favorable as they get for direct pricing — shared geography, integrated economies, meaningful trade flows — the cross still wins.
Herman
Because the EUR pairs have tighter spreads, deeper books, and better hedging infrastructure. The economic incentive always points toward the hub currency. It's not a conspiracy, it's just that liquidity concentrates where liquidity already is.
Corn
There's something almost self-reinforcing about that, isn't there? The dollar pairs are liquid because everyone uses them, and everyone uses them because they're liquid.
Herman
Classic network effect. It's the same dynamic you see with payment systems or operating systems — the value of being on the dominant network increases as more participants join it. In FX, the dollar became the hub in the post-war period partly by design — Bretton Woods explicitly pegged everything to the dollar — but even after that system collapsed in the early seventies, the network effects were already so entrenched that nothing displaced it. The euro was supposed to challenge dollar dominance when it launched, and in some ways it has — EUR/USD is the most traded pair in the world — but the dollar's role as the universal intermediary in cross-pair pricing is essentially untouched.
Corn
When does that calculus actually shift? Because you said direct trading matters — there must be some threshold where a pair develops enough volume to stop being a satellite of the cross.
Herman
There is, and it's worth being specific about what drives it. The clearest cases are regional trade corridors where two economies have such dense bilateral flows that a direct market becomes useful. Thai Baht against Malaysian Ringgit is a reasonable example — ASEAN trade integration has pushed enough volume through that channel that THB/MYR has developed something closer to real independent liquidity. Not deep by major-pair standards, but deep enough that the direct market starts to matter at the margin.
Corn
The geography has to do actual economic work. It's not just proximity, it's transaction volume.
Herman
Right, and you need a few other things too — local banking infrastructure that can handle both currencies, regulatory frameworks that allow direct settlement, and ideally some kind of swap line or central bank arrangement that backstops liquidity in a crisis. When those pieces align, the direct market can develop a genuine footing.
Corn
What does that look like in practice when the swap lines are actually in place? Does that visibly change how the pair trades?
Herman
It can, especially during stress periods. The Fed extended dollar swap lines to a handful of major central banks during the 2008 crisis and again in March 2020 — and what those lines do is guarantee that even if dollar funding dries up in the interbank market, the central bank on the other end can still supply dollars to its domestic institutions. That removes one of the tail risks that makes exotic cross spreads blow out. If you're a market maker and you know the Bank of Thailand can access dollar liquidity through a swap line in a crunch, you're slightly less worried about being caught with Baht inventory you can't hedge. The spread doesn't collapse, but the worst-case scenario gets a bit less catastrophic.
Corn
When they don't align, you see what happened with the Lira in March.
Herman
The Turkish Lira situation is almost a textbook illustration of how cross-rates amplify volatility when things go wrong. TRY/JPY is not a pair most people trade directly. But in March, when you had simultaneous stress in USD/TRY — the Lira under serious pressure — and safe-haven flows pushing USD/JPY down as yen strengthened, those two moves compounded through the cross. TRY/JPY collapsed harder than either constituent pair alone would have suggested, because you were multiplying a weakening currency by a strengthening one, and the direct TRY/JPY market had essentially no cushion to absorb it.
Corn
The cross-rate structure doesn't just transmit volatility, it amplifies it. You're stacking two separate risk events into one number.
Herman
That's the slippage dynamic that catches people off guard. In normal conditions, the cross-rate is efficient — it's pulling from liquid dollar pairs, the arbitrage tether holds, everything is tidy. But when USD liquidity itself starts drying up — which happens fast during a genuine risk-off event — both legs of the cross can move sharply and in adverse directions simultaneously. The bid-ask spread on the exotic pair can widen to levels that make hedging essentially prohibitive.
Corn
Which brings you to the corporate treasurer sitting in Istanbul trying to hedge yen-denominated obligations. Suddenly the instrument they thought they understood has become something much nastier.
Herman
This is the practical consequence that doesn't get enough attention. When a multinational runs emerging market operations — say a European company with factories in Thailand and South Africa — they think of their currency exposure as local. Baht costs, Rand revenues, manage the mismatch. But what they're actually running is hidden USD exposure on both sides. If the dollar strengthens sharply, USD/THB and USD/ZAR both move, and the cross between their two local currencies swings in ways that have nothing to do with Thai or South African fundamentals.
Corn
The dollar is the silent third party in every one of those transactions.
Herman
And the hedging cost reflects that. When banks price a forward contract for an exotic cross, they're essentially pricing two separate dollar hedges and charging you for both, plus the illiquidity premium on top. Corporate treasury teams that understand the underlying structure can sometimes reduce that cost by hedging the dollar legs separately — but most don't, and they pay for it.
Corn
Is there a meaningful difference in how exposed you are depending on whether your two local currencies tend to move together or in opposite directions against the dollar?
Herman
This is where correlation comes in, and it's underappreciated in how cross-rate risk gets framed. If you're running a cross between two emerging market currencies that tend to sell off together during dollar strength — both of them commodity exporters, say, or both of them sensitive to the same risk-off dynamic — then your cross-rate might actually be relatively stable even as both constituent pairs move a lot. The movements partially cancel. But if you've got a cross where one leg is a safe haven and the other is a risk currency, like TRY/JPY, you get the opposite — the two legs compound each other in stress. The Lira falls, the Yen rises, and the cross moves more than either leg alone. Knowing that correlation structure going in is the difference between a hedge that works and one that fails exactly when you need it.
Corn
What about central banks? Because you'd think a central bank has more tools to assert its currency's value than a corporate treasurer does.
Herman
In theory, yes. In practice, if your currency is priced primarily through a dollar cross, your ability to manage that rate is constrained by whatever is happening in the dollar market. A central bank can intervene in USD/THB — buy Baht, sell dollars, defend a level. But if they're not also intervening in USD/ZAR, or if global dollar conditions are moving faster than they can respond, the cross rate they care about can drift regardless. The intervention has to happen in the right market, which is usually the dollar pair, not the direct cross.
Corn
Even the official rate for a minor currency is partly a dollar derivative.
Herman
For most of the world, yes. Which is why Fed policy transmits so far and so fast into places that have no direct relationship with the United States. A rate decision in Washington moves USD/everything, and through those moves, it reshapes every cross pair built on top of them. The local central bank is managing the consequences of someone else's monetary policy as much as their own — leaving traders to navigate that volatility in real time.
Corn
And with all that in mind, the hidden dollar exposure, the amplified volatility, the spread mechanics — what does someone actually do with this information when they're sitting in front of a trading platform?
Herman
The first thing, and I cannot overstate how often this gets skipped, is to decompose the pair before you trade it. If you're looking at a THB/ZAR quote, pull up USD/THB and USD/ZAR separately. Check the spreads on those constituent pairs. Check the recent volatility. Because your execution quality on the cross is going to be bounded by what's happening in both of those markets simultaneously. If USD/ZAR is having a wide-spread day because South African political news just dropped, your THB/ZAR trade is paying for that whether you know it or not.
Corn
The cross is only as clean as its dirtiest leg.
Herman
And the second thing — which connects to the Fed transmission point we just made — is to treat USD liquidity conditions as a leading indicator for exotic cross volatility. When the Fed is in a hiking cycle, or when there's genuine dollar funding stress in the system, that's when cross-rate spreads blow out and slippage becomes material. You want to be aware of that before you're in a position, not after.
Corn
Not the kind of risk that shows up in a standard volatility estimate for the pair itself.
Herman
It won't. The historical vol on THB/ZAR will look relatively calm most of the time, and then suddenly it isn't, because a dollar event happened. That gap between normal-conditions vol and crisis-conditions vol is wider for exotics than almost any other asset class.
Corn
Is there a practical way to monitor that? Like, is there a single thing a trader can watch that gives them a decent early warning?
Herman
A few things, actually. The DXY — the dollar index — is the obvious one, but it's a blunt instrument because it's heavily weighted toward EUR/USD. More useful for emerging market crosses is watching the EMFX volatility index, which tracks implied volatility across a basket of emerging market currencies against the dollar. When that starts climbing, it usually means dollar funding conditions are tightening and exotic cross spreads are about to follow. You can also watch cross-currency basis swaps — specifically the spread on dollar basis swaps for currencies like the Baht or the Rand. When those spreads widen, it's a direct signal that dollar liquidity in those corridors is getting expensive, which is exactly the condition under which your exotic cross becomes much harder to trade cleanly.
Corn
On the platform side — is there anything a trader can actually look for to know whether they're getting a cross-derived price or something with genuine direct liquidity behind it?
Herman
Some platforms do tag this. The terminology varies, but you're looking for labels like composite pricing versus direct pricing, or sometimes synthetic versus native. A composite quote is built from the constituent dollar pairs in real time. A native or direct quote means there's an actual order book for that pair with real resting liquidity. The spread on a composite quote will often be tighter on paper but can gap badly under stress. The direct quote might look wider in normal conditions but behaves more predictably when markets move fast.
Corn
Tighter spread is not automatically better.
Herman
Especially not for exotics. The spread you see in calm conditions on a composite quote is not the spread you'll get when you need to exit a position quickly. That's the number that matters.
Corn
Which is the kind of thing you only learn the hard way, usually at the worst possible moment.
Herman
And look — none of this means you can't trade exotic crosses. It means you need to understand what you're actually trading, which is a compound instrument built on dollar plumbing. Once you see it that way, the risk profile makes a lot more sense.
Corn
Here's the question I keep coming back to, though. We've been talking about this structure as though it's essentially permanent — USD at the center, everything else derived from it. But there's a real conversation happening about whether central bank digital currencies could change that. Direct peer-to-peer settlement between, say, a digital Baht and a digital Rand, without routing through correspondent banking in dollars at all.
Herman
It's interesting to think about. The theoretical case is there — if two central banks issue CBDCs on interoperable infrastructure, you could imagine direct atomic settlement without a dollar intermediary. No correspondent bank, no dollar leg, no cross-rate arbitrage tether. The plumbing just... goes around the hub.
Corn
There are actually live experiments along those lines, aren't there? I seem to remember the BIS running something in Southeast Asia specifically.
Herman
Project mBridge is the one you're thinking of — it's a multi-CBDC platform developed with the BIS Innovation Hub and the central banks of Thailand, Hong Kong, China, and the UAE. The idea is exactly what you described: direct cross-border settlement between participating currencies without routing through a dollar correspondent. They've done real-value pilots, not just proofs of concept. And on a technical level, it works — you can settle a Thai Baht to UAE Dirham transaction directly on the shared ledger. The question is whether that technical capability translates into the kind of market depth and pricing efficiency that would actually challenge how cross-rates are anchored today.
Corn
Does the cross-rate structure collapse in that world?
Herman
I think liquidity still concentrates somewhere. Even if CBDCs enable direct settlement technically, you'd need deep order books on both sides to get efficient pricing. That takes time to build, and it builds where transaction volume already is. The dollar didn't become the hub currency because of a technical constraint — it became the hub because of trade patterns, reserve holdings, and decades of institutional trust. CBDCs can change the rails, but changing the traffic patterns is a much longer project.
Corn
The USD stays as global plumbing even if the pipes get rebuilt.
Herman
For the foreseeable future, yes. Even a seemingly local trade between two minor currencies is, in most cases, a global trade wearing local clothes. That's the thing about the dollar's structural role — it's not just in the big obvious transactions. It's in everything.
Corn
On that note — big thanks to Hilbert Flumingtop for producing this one. And to Modal for keeping our infrastructure running without us having to think about it. This has been My Weird Prompts. If you're listening on Spotify, a review goes a long way — find us there and let us know what you think.
Herman
See you next time.

This episode was generated with AI assistance. Hosts Herman and Corn are AI personalities.