Currencies do not just drift; they sometimes collapse, suddenly and violently, in crises that wreck economies. Economists have developed three 'generations' of models to explain why — each prompted by a crisis the previous generation could not explain. This module covers all three, which together give you a toolkit for diagnosing any currency crisis, including the recurring African ones.
First generation: fundamentals and reserves
Krugman (1979): the inevitable attack
The first-generation model (Krugman 1979, building on Salant-Henderson) explains crises driven by inconsistent fundamentals. A government pegs its exchange rate but runs persistent fiscal deficits financed by money creation. The excess money creation is inconsistent with the peg (it puts downward pressure on the currency), so the central bank must sell reserves to defend the peg. Reserves steadily fall. Rational speculators, foreseeing that reserves will eventually run out and the peg will collapse, attack BEFORE that point — buying up the remaining reserves in a sudden speculative attack that brings forward the inevitable collapse. The crisis is the logical, predictable consequence of fundamentals (the deficit-and-peg inconsistency); the timing is determined by speculators rationally jumping before reserves hit zero. The policy implication: don't peg while running deficits financed by money creation — the crisis is baked in (the fiscal-dominance theme of the Monetary Policy course, on the external side). This fits classic balance-of-payments crises driven by loose fiscal/monetary policy under a peg.
Second generation: self-fulfilling expectations
Obstfeld: multiple equilibria
The 1992–93 European Exchange Rate Mechanism crises (the UK's exit, 'Black Wednesday') did not fit the first generation — the countries attacked did not have obviously inconsistent fundamentals. Obstfeld's second-generation models explain self-fulfilling crises with multiple equilibria. The government weighs the costs and benefits of defending the peg: defending requires high interest rates, which hurt the domestic economy (unemployment, debt service). If markets believe the peg will hold, the cost of defending is low and the government holds — a good equilibrium. But if markets believe the peg will break, they attack, forcing the government to raise rates painfully high to defend; the cost of defending rises so much that the government rationally chooses to abandon the peg — validating the attack. So the crisis is self-fulfilling: the currency was defensible if no one attacked, but the attack itself makes defence too costly, so the expectation of collapse causes the collapse. There is no unique 'fundamentals' answer — the same country can be in the good or bad equilibrium depending on sentiment, which is why these crises seem to strike 'out of nowhere' and are prone to contagion.
Third generation: balance sheets and twin crises
The Asian 1997 model: currency-mismatch crises
The 1997 Asian financial crisis fit neither earlier generation — these were fiscally responsible, fast-growing economies. Third-generation models (Krugman 1999; Chang-Velasco) put the financial sector and balance sheets at the centre. The mechanism: banks and firms had borrowed heavily in foreign currency (the currency mismatch / original sin of the Sovereign Debt course) — their liabilities were in dollars, their assets and revenues in local currency. When the currency depreciated, the local-currency value of their dollar debts exploded, destroying their net worth, causing defaults and a banking collapse — which deepened the recession and the capital flight, causing further depreciation, in a vicious spiral. The depreciation and the financial collapse fed each other. This is the 'twin crisis' — a currency crisis and a banking crisis simultaneously, each amplifying the other through balance sheets. The key vulnerability is not fiscal deficits (first generation) or pure sentiment (second) but the currency mismatch in private and financial-sector balance sheets, which turns a depreciation into a self-reinforcing financial collapse. This is the most relevant model for many modern emerging- and frontier-market crises, including African economies with heavy foreign-currency corporate and bank exposure.
Diagnosing a crisis
The three generations are not rivals but a toolkit — most real crises blend them. The diagnostic questions: Are the fundamentals inconsistent (a peg with money-financed deficits)? — first generation. Is the country in a vulnerable zone where sentiment can tip it either way, with self-fulfilling attack risk? — second generation. Are there large currency mismatches in bank and corporate balance sheets that would turn a depreciation into a financial collapse? — third generation. The 2022–2023 pressures on African frontier markets blended elements: deteriorating fundamentals (debt, twin deficits — first-generation flavour), a sudden shift in global sentiment and risk appetite (second-generation/self-fulfilling, driven by the global financial cycle of module 8), and currency mismatches from the Eurobond and dollar-debt build-up (third-generation balance-sheet risk). Reading which mechanisms are at work tells you both the cause and the cure — fix the fundamentals, restore confidence, and/or reduce the balance-sheet mismatches.
Exercise
Three currencies collapse in the same year. Currency X: the country pegged its rate while running large money-financed fiscal deficits, and reserves had been steadily draining. Currency Y: the country had decent fundamentals, but a regional crisis spooked investors who suddenly attacked, and the government abandoned the peg rather than hold punishingly high interest rates. Currency Z: banks and firms had huge unhedged dollar debts; a moderate initial depreciation exploded their debt burdens, triggering bank failures and a deeper collapse. (1) Identify the generation of crisis for each. (2) Explain why each generation arose to explain a crisis the previous couldn't. (3) For Currency Z, explain the vicious spiral in detail. (4) What different policy lesson does each crisis teach?