Which currency should I invoice in?
Short answer. Invoicing in your own currency moves the exchange-rate risk to your client, which is usually the right default, though a third currency is a common compromise. Either way the currency is only half the decision. The other half is how long the invoice stays unpaid, because that is the window during which the rate can move, and shortening it often does more for you than the currency choice did.
What does invoicing in my own currency actually do?
It fixes your number and leaves your client's number floating.
If you invoice 2,000 in your currency, you are owed 2,000 in your currency whatever the rate does. Your client agreed to a figure in a currency they do not hold, so the cost to them changes between the day they agree and the day they pay. The risk did not disappear. It moved across the table.
Invoicing in their currency does the opposite. Their cost is fixed and yours floats. You will be paid an amount you cannot know in advance in the terms that matter to you.
Neither is inherently correct. What matters is who is better placed to carry the uncertainty, and in most freelance arrangements that is the client, because the client may be better placed to absorb the uncertainty, particularly when the invoice is small relative to their overall costs. That is not automatic: a client who is a sole trader or a small agency may be no better placed than you are.
This is roughly what firms do in practice. Research on invoicing behaviour using firm-level data finds that small exporters that do not import tend to price in their own currency, while larger firms with foreign input costs behave differently. If you buy nothing abroad, your position is closer to that of the small exporter.
Why do people say to bill in your own currency by default?
Because it is the option that removes uncertainty from the side least able to absorb it.
That reasoning holds most of the time and it is why the standard advice is what it is. Two situations weaken it.
The first is competitive. Quoting in a currency your client does not use makes your price harder to compare against local alternatives and puts an unpredictable cost on their side. Some clients will accept that. Some will quietly prefer the supplier who quoted in their own currency.
The second is practical. If a client insists, insisting back over a small invoice can cost you more in goodwill than the exposure is worth. Knowing roughly what the exposure is worth is what lets you decide that quickly, and that is arithmetic rather than judgement.
Is there a third option?
Yes, and it is more common than the binary framing suggests: invoice in a currency belonging to neither of you.
Third-currency invoicing is common in international trade, particularly through dominant vehicle currencies such as the US dollar. Roughly 40% of goods trade is invoiced in dollars against a United States share of that trade of around 10%, illustrating how often the dollar is used as a vehicle currency between parties who are not American. If you are a freelancer in one country billing a client in another and both of you reach for dollars, you are using the same kind of vehicle-currency arrangement that is common in international trade.
The reason is more than habit. Firms tend to price in whichever currency keeps their own prices most stable, and for many that is neither the local currency nor the client's. It is worth asking the same question of yourself: in which currency is the number you charge least likely to need revising?
The trade-off is that a vehicle currency leaves both sides with some exposure rather than concentrating it on one. That can be easier to agree than asking your client to take all of it. And if you keep the proceeds in that currency rather than converting on arrival, you also end up holding a foreign balance, which has its own cost.
How long am I actually exposed?
From the moment you agree a price to the moment the money converts. That is longer than most people picture.
The window is not the transfer time. It starts when you fix a number and ends when the currency actually changes hands. On a net-30 invoice paid a week late, with a couple of days for the payment to land and clear, you have been exposed for close to six weeks on work you may have finished before you invoiced.
That framing changes what you can do about it. The currency choice is binary and often not fully yours. The length of the window is a commercial term you negotiate all the time without thinking of it as currency management.
Does halving my payment terms halve the risk?
No, and the actual relationship is worth knowing.
If you model exchange rates as a random walk, which is the standard benchmark in this area, the standard deviation of the rate change grows with the square root of time rather than in proportion to it. It is the same relationship that governs splitting a conversion into parts, running in the opposite direction. On that model, cutting the window in half does not cut the spread of possible outcomes in half.
| Payment window | Modelled variability, relative to 90 days |
|---|---|
| 90 days | 100% |
| 60 days | 82% |
| 45 days | 71% |
| 30 days | 58% |
| 15 days | 41% |
Read it in both directions. Moving from net-90 to net-45 removes less variability than it looks like it should, so do not oversell the benefit to yourself. But moving from net-90 to net-15 cuts it by nearly 60%, a larger effect than most currency choices produce, and it also gets you paid sooner.
These figures describe modelled variability under that assumption. They are not a forecast of what you will lose, and real rates do not behave exactly this way. The table is for ranking your options, not for predicting an outcome.
What about a deposit?
A deposit is the strongest tool in this list, and it is not usually thought of as a currency measure.
Money paid at signing carries no exchange-rate exposure at all, because there is no gap between agreeing the number and converting it. Split an engagement 50% upfront and 50% on delivery in 45 days, and half your invoice has zero exposure. The blended exposure is half what it would be billing the whole amount at the end.
That can reduce exposure more sharply than a comparable adjustment to the payment window alone, and unlike the currency question it rarely requires a difficult conversation. Deposits are normal commercial practice for reasons that have nothing to do with currency.
Can I fix the rate instead?
You can fix a rate in the contract, which is different from fixing it in the market.
A clause that sets the exchange rate for the engagement at the rate on the day of signing gives both sides a known number. It does not remove the risk from the world; it allocates it, the same as choosing a currency does. What it adds is predictability for both parties, which is often easier to agree than asking your client to absorb everything.
Two things to get right if you use one. Name the source and the moment, so "the mid-market rate published by X at close on the date of this agreement" rather than "the exchange rate on the date of signing". And set a tolerance band, so the clause only bites if the rate moves beyond some threshold. Without a band you will be renegotiating over rounding.
How do I decide, in practice?
Four questions, in order of how much they usually move the outcome.
- Can I take a deposit? If yes, this does more than everything below it combined.
- What are my payment terms, and can they be shorter? Net-15 with a deposit leaves very little window at all.
- Whose currency, if the client has no strong preference? Yours, by default. If neither of you is comfortable, a vehicle currency splits the difference.
- Is the exposure large enough to negotiate over? Apply your budgeting buffer to the invoice amount. If the answer is a rounding error on the engagement, take whichever currency the client prefers and spend the goodwill elsewhere.
The last question is the one that saves the most time. Most invoices are small enough and short enough that the currency choice is worth less than the conversation about it. Knowing that with a number rather than a shrug is the point.
Sources and further reading: Federal Reserve Board on the random walk as the standard benchmark in exchange rate modelling; Amiti, Itskhoki and Konings via NBER on how firm characteristics determine invoicing currency choice; CEPR on the share of global trade invoiced in dollars and the dominant currency paradigm.
This article describes how invoice currency and payment terms affect exchange-rate exposure. It is not financial, tax or legal advice, and contract terms should be checked against the law that applies to you. The variability table is the output of a simplified model, not a prediction of results.