Curriculum·G503 Risk Systems and Position Limits·about 34 min

Concentration: the position you cannot exit

By the end of this lesson you can

  • →Explain that a concentrated position cannot be exited without moving the price against itself
  • →Describe how a National Australia Bank options desk built a concentrated bet it could not close and hid the losses
  • →Reason that the difficulty of exiting is part of a position's risk, not separate from it
  • →Use limits to keep a position small enough to actually close

Graduate · enrolled learners

This lesson opens with National Australia Bank, 2004.

What happened
Four currency options traders at National Australia Bank positioned the bank's options book for the US dollar's 2003 fall to reverse and for volatility to settle, and as the market moved against them they did not close the positions; they kept them, added proprietary trades that lost more, and let the positions deteriorate for three months, making them larger and harder to exit. To buy time, they hid the growing losses by entering fictitious one-sided internal trades in the window between the end-of-day processing of the trading system and the reconciliation of internal deals, and by exploiting known weaknesses in how the bank revalued currency options, until the losses were discovered and announced on 13 January 2004, first as 180 million Australian dollars and, by 27 January, after revaluation and expected close-out costs, as 360 million Australian dollars, about 260 million US dollars. The position had become concentrated enough that closing it would have moved the market against them and crystallized the loss, so every day they held it rather than exit, and holding it meant hiding it. Nothing exotic was needed; the desk built a bet too big to get out of, then chose concealment over the loss that exiting would reveal, and the concentration was what turned a bad trade into a trap.
The decision point
A concentrated position, one large relative to the liquidity of its market, cannot be exited without moving the price against itself, so the difficulty of getting out is part of the position's risk, not a separate concern, and a position large enough that closing it would crystallize a big loss becomes a trap the holder is tempted to hide rather than exit. National Australia Bank is the case: an options desk built a concentrated bet that the US dollar's fall would reverse it could not close without moving the market and taking the loss, held on and added trades instead, and concealed about 260 million US dollars until it was found. This completes the risk course's picture: a limit must bind, a measure must see, the records must be honest, and, here, a position must stay small enough to exit. Concentration is dangerous precisely because the exit is where the loss is realized, so a position you cannot exit without moving the price is one whose loss you cannot take cleanly, which makes holding and hiding feel easier than closing, exactly the choice the NAB desk made. The risk of a position therefore includes its exit: a small position can be closed at roughly the market price, while a concentrated one moves the price as it is sold, so its true risk is larger than its mark suggests and its loss is worse the moment it is realized. So the discipline is to treat exit liquidity as part of risk, to size positions so they can be closed without moving the market against you, and to use limits that cap concentration and not just directional exposure, because National Australia Bank shows that a position too big to exit is a position that invites concealment, and the concentration itself, not any exotic instrument, is what turned a losing trade into a hidden, growing trap.
Recorded loss
$260,000,000

What you will be able to answer

  • →What happened at National Australia Bank (2004)?
  • →What is true of a concentrated position?
  • →Why does a position too big to exit invite concealment?
  • →How should exit liquidity shape position sizing and limits?

Orientation and Year One are open: anyone can read them without an account. From Year Two onward the lessons are for enrolled learners, because progress through the later years only means anything if it is tracked against a record.

It is free. We do not sell the list and there is nothing to buy at the end of it.

Sources and review

Confidence high·Volatility low·Reviewed 2026-09-18·Owner unassigned

Contested

The 360 million Australian dollars is NAB's restated loss of 27 January 2004, as recorded in APRA's report of 23 March 2004 (which NAB filed with the SEC); it rose from the 180 million announced on 13 January because of revised revaluation rates and an allowance for expected close-out costs. The US dollar figure follows Euromoney's conversion, 263 million dollars.

APRA attributes the loss first to the traders' collusion and then to lax line management, poor adherence to limits (VaR excesses on the desk were switched off for large parts of two years), missed audit and counterparty warnings, and governance that did not function. It does not frame the case as one of exit liquidity: reading the concentrated, hard-to-close position as the trap that invited concealment is this lesson's interpretation, supported by APRA's note that the desk's exposures were concentrated among a few interbank counterparties and that close-out costs were part of the restated loss.