Currency Crises and Speculative Attacks
A speculative attack is a coordinated bet that a central bank will run out of reserves defending its currency peg — and because the bet itself accelerates the reserve drain, it can become self-fulfilling regardless of the peg's original fundamentals.
Prerequisites: Currency Pegs and Managed Floats, Central Bank FX Intervention
George Soros's fund is said to have made about $1 billion betting against the British pound in 1992, forcing the Bank of England out of a peg it had defended for two years. That trade wasn't a lucky guess — it was a bet with a specific, calculable trigger: the Bank of England's reserves and its willingness to raise interest rates had a limit, and once traders concluded that limit had been reached, selling the pound became close to risk-free. That is a speculative attack: a wave of traders selling a pegged currency simultaneously because they believe the peg is about to break, which itself makes the peg more likely to break.
A speculative attack works because defending a peg costs reserves, and reserves are finite and public knowledge. Once enough traders believe a central bank is close to its limit, selling the currency accelerates the reserve drain directly — turning a belief about the peg's weakness into the cause of its collapse, whether or not the peg was actually unsustainable to begin with.
The mechanics of the attack
Classic models (Krugman, 1979) show that an attack doesn't wait until reserves literally hit zero — it happens the moment the market calculates that reserves will hit zero on the current trajectory, because nobody wants to be the last one holding the pegged currency when it breaks:
In words: reserves fall at a steady rate as the central bank defends the peg, and any trader can extrapolate that line forward to estimate when it hits zero. Once that date looks close, the rational move is to sell now, before the rush — which pulls the actual collapse date forward, often by months or years, compared to a naive projection of the outflow rate alone.
Worked example
A central bank has $30 billion in reserves, losing $1 billion a month defending its peg — an 30-month runway on the naive projection. A hedge fund calculates that the country's current account deficit alone requires the bank to sell $1.2 billion a month just to keep pace, meaning the true runway is under 25 months, and that political constraints make raising rates to defend the peg (which would fight a recession) unlikely. The fund starts shorting the currency forward. As other funds run similar numbers and pile in, monthly outflows jump to $4 billion, cutting the real runway to under 8 months — and the central bank, facing an accelerating drain it can now see has no floor, often abandons the peg within weeks rather than months, well before the original 30-month estimate.
What this means in practice
Quants and macro funds watching for an attack track reserve trends, import cover (months of imports reserves could pay for), and the political cost of the interest-rate hikes needed to defend the peg — not the exchange rate itself, which stays artificially flat until the moment it doesn't. Central banks, aware of this, sometimes defend pegs past the point of pure economic sense simply to preserve credibility for the next crisis.
A currency crisis is not proof the original peg was doomed by fundamentals. Self-fulfilling attacks can break pegs that might have survived indefinitely absent the attack — the belief and the outcome are entangled, which is exactly why these episodes are so hard to predict in advance and so obvious in hindsight.
Related concepts
Practice in interviews
Further reading
- Krugman, 'A Model of Balance-of-Payments Crises', Journal of Money, Credit and Banking (1979)
- Obstfeld, 'The Logic of Currency Crises', NBER Working Paper 4640 (1994)