Modelling foreign exchange gains and losses

how to model foreign exchange


Andrew Berkley


06 Oct 2021




Six steps to foreign exchange happiness

If any financial modelling topic requires clear thinking then it is foreign exchange (“forex”). That’s because it calls upon domain knowledge from at least three different areas. You need to have a working understanding of accounting (noting that accounting standards in in respect of forex are far from simple), a passing knowledge of treasury and a dash of economics. Good financial modelling techniques are taken as read.

I teach how to model foreign exchange transactions frequently and it’s often a tough call. That’s because there’s a lot of detailed background information to go through. A few years ago, I came up with “six steps to foreign exchange happiness” – my best attempt to build up the knowledge in layers. And that’s what I’d like to share with you now.


1. Decide on a foreign exchange quotation convention – and be clear in the units that you display.

There is an ambiguity in stating an exchange rate as “GBP USD” (for example). Does that mean a variable number of USD for every one GBP? Or is it the other way round: variable GBP for one USD? 
In F1F9, we use the base / variable convention: the base currency is on the left; the variable currency is on the right. That is what you’ll find if you search for GBP / USD”: a variable number of USD for every GBP.

Which means that the units in your model should be “USD” or even “USD per GBP”.

The 3-letter currency codes that I am using are themselves the subject of an international standard (ISO 4217). Given the work that must have been done to get global agreement, it seems sensible to take advantage of it.

2. Agree on the types of currency that you are working with.

There are three types of currency that a financial modeller needs to find:

Transactional currencies: there can be numerous transactional currencies. For example, currencies in which revenue is received (GBP, USD, AED and EUR for F1F9). And other transactions may happen on the cost side: INR for F1F9. That is 5 separate transactional currencies that would need to be modelled in any forecasts relating to F1F9.

Functional currency: there is only one functional currency – and it is up to you what that currency should be. All transactional currencies need to be converted into that single functional currency – so choose wisely. Most financial modellers work with the most used transactional currency and adopt that as their functional currency. Why? Because that results in the least work.

Presentational currencies: organisations may choose to present their forecasts in different currencies – so there can be many presentational currencies. A financial modeller wishing to minimise work can choose a presentational currency that matches the functional currency. That is often sufficient when the primary concern is, for example, testing a project’s ability to service debt against worsening exchange rates.

Important note: there are differences in how accounting standards recommend you convert from:

– transactional currencies to functional currencies; and
– functional currencies to presentational currencies

My advice to financial modellers: don’t try to bypass the functional currency, however tempting it might appear.

3. Agree on the timing of flows.

When converting from transactional currencies to a single functional currency, accounting standards start with the balance sheet. Put simply, the guidance is to convert monetary balances (debt, working capital) using a closing rate (the current – or spot – rate applicable at the date of the balance sheet).

When choosing rates to convert flows, accounting standards offer other options – including an average rate (calculated over the period of the income statement, for example). While this in some cases gives a more exact answer, that accuracy is not always significant. It can also generate plenty of modelling work when it comes to calculating gains and losses arising.

If you can make the simplifying assumption that both flows and balances will be converted at a closing rate (matching the date of the balance sheet), then that will simplify the modelling.

4. Convert the balances.

Take the forecast balances expressed in their transactional currencies and convert them to equivalent balances expressed in the functional currency. That will require a forecast of future spot exchange rates.

The four way equivalence model is a good place to start if you would like to better understand the relationship between interest rates, inflation rates and foreign exchange rates. Check out the purchasing power parity theory and the international Fisher effect.

5. Convert the flows.

Using the same set of forecast closing rates, calculate the flows (expressed in their equivalent functional currency). Remember: your assumption is that flows and balances are converted using the same closing rate (assuming flows happen on the same date that balance sheets are prepared).

That is something that’s quite realistic for, say, senior debt (when debt service is paid according to a fixed schedule and the next period’s interest is calculated from the date of the last debt service). It is less realistic for working capital balances – but then again working capital tends to be far less sensitive when running “what if?” analysis. There is a balance to be struck between accuracy and minimising unnecessary modelling work.

6. Calculate the gain / loss arising from foreign exchange.

Gains or losses that have yet to have an impact on cash flows (known as “unrealised gains or losses”) arise when putting balance sheets together. That’s because exchange rates will have changed between balance sheet dates. A year ago you may have owed 10m; now, because of worsening exchange rates, you owe 15m. That’s an unrealised loss of 5m that needs to be recognised in the income statement.

Note that the loss has arisen on a closing balance. Closing balances are made up of three things:

A beginning balance; plus

An upward flow; less

A downward flow

If you assume that the timing of flows is end of period (and many financial models make this simplifying assumption), then it is justifiable for you to apply a closing rate to both balances and flows. In turn, this leads to exchange rate movements on the flow becoming irrelevant when converting the closing balance.

And we can conclude that if closing rates are used, only changes in the beginning balance from one period to the next arising from movements in exchange rates will give rise to foreign exchange gains or losses in your financial model. Gains or losses may be calculated by converting the beginning balance at the closing rate and deducting the same beginning balance converted at the prior period’s closing rate.

Forex is just one of many topics we’ve incorporated into custom courses for in-house clients. If you would like to learn more about how we can tailor training for your team you can watch our short video and book a call with me to talk through your requirements.

Andrew Berkley
With a background in business education and financial advisory work, Andrew leads financial modelling training at F1F9. He has been with F1F9 since 2013.