Market risk is the one exposure every trading company believes it measures. The number is not wrong - it has a boundary. It begins when a trade is captured, it is struck at a moment, and it is marked at a mid, and the exposure it measures respects none of those three limits.
Summary
Market risk is the one exposure every trading company believes it measures. There is a number for it, it is produced daily, and it is the number the board asks about. The difficulty is not that the number is wrong. It is that the number has a boundary — it begins when a trade is captured, it is struck at a moment, and it is marked at a mid — and the exposure it is measuring respects none of those three limits.
This article sets out where price exposure lives when it is not yet on the risk report: in commitments that have been made but not yet booked, in the gap between the moments at which the book is measured, and in the difference between the price you are marked at and the price at which you could actually get out. It is written for the people who own the consequences: traders whose position is larger than their book says, the risk and middle office who produce the number and know what it excludes, and the commercial side who create exposure by saying yes to a customer.
If you want the mechanism that costs the most and shows the least, start with Exposure begins when you commit, not when someone books it. If your book is already captured cleanly and you want the part that only appears under stress, start with The mark is a mid, and the exit is a spread.
Exposure begins when you commit, not when someone books it
A position system measures trades. A trading business takes on price exposure by making commitments, and a commitment becomes a trade only after somebody has entered it.
The gap between those two events is not a data quality problem to be tightened up. It is a structural feature of how commercial businesses work, and several ordinary activities live inside it.
A firm offer is a written option that you have given away. When a desk quotes a fixed price to a customer and holds it open until the end of the day, the customer holds the right — not the obligation — to trade at that price for as long as the offer stands. If the market moves in the customer's favour, the offer is lifted; if it moves the other way, it is not. That asymmetry is the definition of an option, and it does not stop being one because no premium changed hands and no ticket was written. For the period the quote is live, the exposure is real and it is on nobody's report.
Volume committed under a term contract is exposure before it is a trade. A contract to supply a set quantity each month, at a formula rather than a fixed price, creates an obligation the day it is signed. Depending on how the system is set up, the individual months may only appear as trades when they are nominated, scheduled or priced. Everything between signature and that moment is a position held under contract and reported, at best, as a memo.
Inventory is a price position with no trade behind it. Material in a tank was bought at a price that has since stopped being the market's price. Whether that shows up as exposure depends on how the inventory is valued and whether the valuation feeds the same report as the trades. Where it does not, a business can be substantially long and show a flat book.
Optionality written into physical contracts is exposure that was never traded. Quantity tolerances, the right to choose a delivery window, the right to substitute a grade or a delivery point — each is a choice held by one party, and every choice held by the other party is a short option held by you. These are agreed in the contract, exercised by email, and recorded in the position system rarely.
The consequence is worth stating plainly, because it runs against the intuition that a clean book means a measured book: the completeness of your price risk number depends on the boundary of your trade capture process, and that boundary was drawn for accounting reasons rather than risk reasons. A trade is captured when it is certain enough to be booked. Exposure begins when it is probable enough to move money.
The number is a snapshot of something that moves continuously
A market risk figure is struck at a moment. The exposure it describes exists at every moment, including all the ones between two measurements.
Most risk reporting is end-of-day: positions as at a close, marks as at a close, limits tested against that. This is a sensible convention and it is not the problem. The problem is what it licenses people to believe.
A limit tested at the close is not a limit. If the measured figure is inside the limit at 17:00 on Monday and inside it again at 17:00 on Tuesday, nothing has been established about Tuesday. A book can go substantially through a limit intraday and return, and the report will show two compliant days. Whether that matters depends on how the business trades; for a desk that turns over during the session it matters a great deal, and the report is silent on it by construction.
The gap is largest exactly when it matters most. Intraday range widens when the market is moving, which is when positions are being changed and when the difference between two snapshots is least representative of what happened in between. The measurement is at its least informative in precisely the conditions it exists to cover.
And the reporting lag is a second, separate gap. Even where the snapshot is honest, it describes a book that is already some hours old by the time it is read. For exposures whose governing facts arrive from outside the trading system — an operational delay that moves a pricing window, a nomination that fixes a volume, a customer accepting yesterday's offer this morning — the lag is longer than a day, and it is not visible as a lag. It is visible as a position that appears to have changed overnight for no reason.
None of this argues for measuring continuously, which for most businesses would be an expensive answer to a question they have not yet asked. It argues for knowing which of your exposures are stable between snapshots and which are not, because the report treats them identically and they are not identical.
The mark is a mid, and the exit is a spread
A position is marked at a price nobody has offered you. Closing it requires a price somebody has.
Marking to market normally means marking to the middle of a quoted spread, or to a settlement price derived from one. That is the right convention for comparing today with yesterday, and it is the wrong number for answering "what would we get if we had to be out of this by Friday." The difference between the two is a real cost that does not appear as a position.
Three things make the gap larger than it looks.
The spread is not a constant. It widens when liquidity thins — around holidays, around expiries, in contract months further out, and in any period when the market is repricing quickly. The mark keeps updating smoothly through all of that, because a mid can be computed from quotes that nobody is prepared to trade in size.
Size is priced separately. A quoted spread applies to a clip. A position several multiples of that size does not exit at the quoted spread, and the amount by which it does not is a function of how badly you need to be out. This is the part that is hardest to hold in mind when the position is being built, because at that point you are a price taker on the way in and the market is accommodating.
And the exposures that are hardest to exit are usually the ones that were hardest to hedge in the first place. A liquid benchmark leg can be closed at something close to its mark. A position in a thinly quoted differential, a distant contract month, or a physical grade with few natural buyers cannot. The residual risks that a hedging programme leaves behind are systematically the illiquid ones, which means the portion of the book that is marked least reliably is the portion the hedge did not cover.
The useful question is not "what is this worth" but "who would take the other side, and at what." For a large part of a physical book, the honest answer is that the counterparty would have to be found rather than looked up — and a valuation that assumes otherwise is describing a market that would have to be created before it could be used.
Hedged is a shape, not an amount
A hedge does not remove price exposure. It exchanges one exposure for another, usually smaller and usually better understood, and the residual is a position in whatever the hedge did not track: the difference between the instrument and the thing, the difference between their dates, and the difference between their liquidity.
This is the same mechanism that makes a benchmark hedge leave a differential open, and it is worked through in detail for physical crude in Crude Oil Trading: Why the Barrel You Hedge Is Not the Barrel You Own. The general point is the one that belongs here: a book described as hedged has had its risk reshaped, and the report that says "hedged" is usually reporting the amount rather than the shape.
Where price exposure lives before it reaches the risk report
| Firm offer outstanding | Term contract signed, month not yet nominated | Inventory held | Contract optionality granted | |
|---|---|---|---|---|
| What creates it | A quote held open for a customer | Signature on a supply agreement | Material bought and not yet sold | A choice given to the counterparty |
| Why it is exposure | You hold a short option; the customer decides | The obligation exists; the price does not | Replacement cost has moved away from cost | Every choice they hold is a short option you wrote |
| Where the fact first exists | The trader's outbox or the quoting screen | The contract, and the legal file | The tank gauge and the purchase invoice | A clause, and later an email exercising it |
| When the book usually learns | When the offer is lifted | At nomination or at pricing | At month-end valuation, if inventory is revalued | When the choice is exercised |
| What the report shows meanwhile | Nothing | A memo line at best | Depends entirely on the valuation policy | Nothing |
| The question to ask | How long do our quotes stay live, and how big can they be | Which months are obligations and which are trades | Does inventory revaluation reach the same report as trades | Which of our contracts contain a choice we do not hold |
The bottom row is the one to work through with your own book. None of these are exotic instruments; all four are ordinary commercial practice. What decides whether they are managed is whether anyone has drawn the line between "commitments the business has made" and "trades the system has captured", and looked at what sits between them.
Questions people ask about this
Is market risk the same thing as price risk?
In practice the two terms are used for the same exposure: the risk that the value of what you hold, or have committed to, changes because market prices move. Market risk is sometimes used more broadly to include the interest rate and currency components of the same book, which are separate mechanisms with their own articles. If you are trying to decide which report you need, the more useful distinction is not between the two words but between exposure that arises from a captured trade and exposure that arises from a commitment that has not been captured — because those two are usually produced by different systems, and often by different departments.
Our value-at-risk number is inside its limit. Are we within appetite?
You are within the limit on the measured book at the measurement time, which is a narrower statement than it sounds. Three things sit outside it: exposure created by commitments not yet captured as trades, movement between one snapshot and the next, and the difference between the marks used and the prices at which the positions could actually be closed. None of those makes the number wrong. All three mean the answer to "are we within appetite" requires knowing what the number excludes, and that list is stable enough to be written down once and reviewed rather than rediscovered each time.
How can we be exposed to a trade we have not done?
Because you have already given someone else the right to decide. A firm price held open for a customer is an option they hold and you wrote — they will take it if the market moves their way and leave it if it does not. The same shape appears in physical contracts as a quantity tolerance, a delivery window the counterparty declares, or a right to substitute a grade. In each case the decision belongs to the other party, and until they make it, the outcome moves with the market on your side of the trade rather than theirs.
Does hedging remove price risk?
It replaces it. A hedge tracks the thing you own well enough to cancel most of the movement, and what remains is the part it did not track — a different grade, a different location, a different set of dates, a different depth of market. That residual is usually smaller and is often better understood, which is what makes hedging worth doing. But it is a position, and a book that reports itself as hedged without reporting what it is left with has described the size of the operation rather than the state of the risk.
Why does our profit and loss move on a day when our position did not?
Usually because something other than the position changed: the marks, the shape of the curve between the dates you are exposed to, or the relationship between the instrument you hedged with and the thing you hedged. A second and less obvious cause is that the position did change and the change had not yet reached the report — a customer lifting yesterday's offer, a nomination fixing a volume, an operational event moving a pricing date. Both look identical on the report, and telling them apart requires knowing when each fact first existed somewhere in the company.
Can we manage market risk in spreadsheets?
Many companies do, and for a book whose terms one person can hold in their head it works. What tends to fail first is not the calculation but the boundary: a spreadsheet is maintained by whoever maintains it, and the commitments that have not become trades are precisely the items nobody has been made responsible for entering. The threshold is not a volume. It is the point at which the person who knows which quotes are still live is no longer the same person who updates the position file.
What should we measure first if we are starting from nothing?
Start with the inventory of what creates exposure, not with a risk model. Write down every way your business becomes exposed to a price move — trades, live quotes, signed term volume, inventory, and every contractual choice held by a counterparty — and mark for each one whether it reaches the risk report, and how late. That list is short, it can be built in a week from people who already know the answers, and it will usually show that the largest available improvement is in coverage rather than in the sophistication of the calculation.
Our exposure is all in one liquid benchmark. Do these problems apply to us?
The mark and exit problem applies least to you, which is a genuine advantage. The other two still apply. A liquid benchmark position that is entered on the basis of a customer commitment has the same gap between commitment and capture as any other, and a snapshot-based limit is a snapshot-based limit regardless of what is being measured. It is worth checking the second one specifically: liquidity is what makes intraday turnover practical, so the businesses with the least mark-to-exit risk are often the ones with the most movement between two measurements.
Where this lands in a trading system
Nothing above is an argument for buying software. It is an argument about which facts have to exist, where, and how soon after they become true. The conclusions do map onto system capabilities, and it is worth being specific about which ones and about where the mapping stops.
The first conclusion is that trades and the instruments used to hedge them have to be held together before a residual can be seen at all. That is the job of a CTRM/ETRM system providing complete trading lifecycle management for commodity traders, integrating physical trade, financial hedging, risk control, and settlement in one platform; Time Dynamics' Fusion is one, and the parts of it that bear on this article are that complete physical trade lifecycle management and execution and advanced trading management for futures and derivatives markets sit in the same place, with comprehensive risk assessment and monitoring capabilities applied across both.
The second conclusion is that a large share of what creates price exposure is not in that system and will not be moved into it: quotes in an outbox, obligations in a signed contract, tank gauges, and the email in which a counterparty declares a choice. That is a different problem, and it is the one X-Ray addresses — a non-invasive data processing and analysis platform, whose collection toolkit XDK performs non-invasive automated data collection from databases, Excel files, and web interfaces, and which is built to collect data without disrupting existing systems or workflows. The last property is the one that matters here, because the files holding your live quotes and your open contract terms are working, and a plan that starts by replacing them will not survive its first busy week.
The third conclusion is that the useful view is not one number but the same book re-cut several ways — by what is captured against what is committed, by what is liquid against what is not. X-Sheet does one-click generation of personalized reports with Excel-like interface. The alerting component, X-Eagle, is described as real-time risk alerts and monitoring for trade finance field warnings; whether a threshold of that kind is pointed at the gap between committed and captured volume is a configuration question rather than something a platform decides for you, but a ceiling on a quantity that moves during the day is an ordinary shape rather than an exotic one.
None of this sets your risk appetite, and none of it decides which exposures are worth carrying. It will not tell you whether a differential is cheap, it will not decide how long your quotes should stay live, and it cannot value a position in a market where the buyer would have to be found rather than looked up. It will not read a contract and tell you that the tolerance you granted is expensive. What a system can settle is narrower and duller than the problem in this article: where each of these facts is captured, how soon after it becomes true, and whether the person who needs it can see it without asking the person who created it. That is a real part of the gap. It is not the whole of it, and it is worth being clear about which part you are buying.
A note on the evidence in this article
This article contains no statistics, and that is a deliberate choice rather than an omission. The observations behind it — which parts of a trading business create price exposure before a trade exists, and where the governing facts are held — come from material we are not in a position to publish as a citable figure. A number we could not source would make the argument look stronger and would make it worth less.
So the argument here is structural instead. Each claim is made from the mechanism, and each is stated in a form you can check against your own operation rather than take on our authority. The tests proposed above — how long our quotes stay live and how large they can be, which months are obligations rather than trades, whether inventory revaluation reaches the same report as the trades, which of our contracts contain a choice we do not hold — are the intended way to use it. If your own book contradicts what is written here, your book is the better evidence.