Floating vs Fixed Exchange Rates

A spectrum, not a binary

"Fixed vs floating" is shorthand for a continuum. The IMF's Annual Report on Exchange Arrangements and Exchange Restrictions classifies regimes into ten categories, ordered from hardest fix to freest float:

1. No separate legal tender (dollarization, e.g. Ecuador, El Salvador, Panama).
2. Currency board (Hong Kong since 1983; Bulgaria since 1997).
3. Conventional peg within ±1% (Saudi Arabia at 3.75 SAR/USD).
4. Stabilized arrangement.
5. Crawling peg (rate adjusted in pre-announced steps; Nicaragua).
6. Crawl-like arrangement.
7. Pegged within horizontal bands (e.g. ERM ±2.25% pre-1992).
8. Other managed arrangement.
9. Floating (intervention rare, no target — Mexico, India, Brazil).
10. Free floating (no FX intervention except for orderly markets — U.S., U.K., Japan, eurozone).

Roughly 40% of IMF members use some form of peg, 40% manage their float, and only ~20% truly free-float.

The impossible trinity

Robert Mundell (1960s) and Marcus Fleming (1962) formalized the constraint. Pick any two corners of the triangle and you forfeit the third:

• Fixed rate + open capital + independent rates → impossible. Capital arbitrages your interest-rate target away. Try to keep rates above world levels and money pours in, blowing the peg upward; below, money flees and reserves drain.
• Fixed rate + open capital, no independence. Hong Kong's choice. The HKMA imports U.S. monetary policy through the peg.
• Fixed rate + independent policy, capital controls. China pre-2015. The PBoC could set domestic rates because the capital account was closed.
• Floating + open capital + independent policy. Most advanced economies post-1973.

Worked example: Hong Kong's currency board vs Mexico's float

Two open economies, two opposite choices.

Hong Kong, peg at HK$7.80/USD (since 1983). The Hong Kong Monetary Authority operates a strict currency board: every Hong Kong dollar in circulation is backed by U.S. dollars in the Exchange Fund. To inject HKD, the HKMA must buy USD; to withdraw HKD, sell USD. The peg has survived (a) the Asian financial crisis of 1997-98, when George Soros and others attacked it, costing the HKMA ~$15 billion of intervention; (b) the 2008 crisis; (c) the 2020 protest-and-pandemic year. Inflation tracks U.S. inflation with a small wedge for housing. The cost: Hong Kong has no domestic monetary policy. When the Fed hikes, Hong Kong hikes — even if the local economy is in recession.

Mexico, free float (since 1994). After the 1994 Tequila crisis broke the crawling peg, Mexico adopted inflation targeting (formalized 1999) and let the peso float. The peso ranges widely — 18 to 24 per USD across 2019-2024 — but Banxico can set rates to fight Mexican inflation without watching the FX screen first. The 2022-2023 hiking cycle (450 bps to 11.25%) responded to Mexican CPI, not the Fed alone. Cost: more day-to-day FX volatility, more import-price pass-through, deeper dollar-denominated debt risk.

Floating vs fixed: full comparison

Property	Floating	Fixed (peg / currency board)
Price discovery	Continuous market	Administered, defended by intervention
Monetary independence	Yes — set rates for domestic goals	No — rates follow anchor country
Reserve requirement	Modest (smoothing only)	Large; currency board needs 100% backing
Adjustment to shocks	Currency moves absorb the shock	Domestic prices/wages must adjust (slow, painful)
Speculative attacks	Limited — there's no peg to break	Recurring risk; ERM 1992, Asia 1997, Argentina 2001
Inflation discipline	Needs domestic anchor (inflation target)	Imported from anchor country
Trade and FDI predictability	Lower — hedging required	Higher within the peg zone
Cost of crisis	Frequent small adjustments	Rare but catastrophic re-pegs / collapses

When a fix is worth it

• Small open economies with a dominant trade partner. Hong Kong's trade and finance are dollar-denominated; pegging removes a transaction cost.
• Post-hyperinflation credibility imports. Argentina 1991-2001, Bulgaria 1997, the Baltic states pre-euro all used hard pegs to break inflation expectations.
• Currency unions. The euro is the limit case — eleven and now twenty member states gave up monetary independence in exchange for a deep, deep peg.
• Oil exporters. Saudi Arabia, the GCC peg to the dollar because their export receipts already are dollars.

When floating wins

• Diversified economies with idiosyncratic shocks. A country whose business cycle differs from its anchor's needs its own monetary policy.
• Credible central banks. Inflation targeting works only with central-bank independence and good communication. NZ pioneered it in 1990; ~40 countries have followed.
• Open capital accounts. By the trilemma, mobile capital + monetary independence requires floating.
• Commodity exporters with floating partners. Australia, Canada, Norway use floats to absorb commodity-price swings — the rate moves first, real adjustment is gentler.

Variants and hybrids

• Crawling peg. Rate adjusts in pre-announced increments to keep up with inflation. Nicaragua, historically Brazil and Chile.
• Target zone. Defend a band, allow movement inside. The European ERM (1979-99) used ±2.25% (later ±15%) bands.
• Managed float. No public target, but the central bank intervenes opportunistically. India's RBI, Brazil's BCB.
• Dollarization. Adopt a foreign currency as legal tender — a one-way irreversible peg. Ecuador (2000), El Salvador (2001), Zimbabwe (2009-19).
• Currency union. Multiple sovereigns share one currency. The euro, the CFA franc zones, the Eastern Caribbean dollar.

Counterarguments and qualifications

• "Floats give monetary independence." Hélène Rey (2013) argued there's actually a dilemma, not a trilemma: global financial cycles driven by U.S. monetary policy and risk appetite spill into floating economies regardless. Independence is partial, not absolute.
• "Pegs are inflation discipline." Only as credible as the commitment. Argentina's convertibility law (1991) was constitutional but still broke in 2002 when fiscal stress overwhelmed it.
• "Markets price floats efficiently." Floating rates exhibit excess volatility relative to fundamentals (Meese-Rogoff, 1983: random walks beat structural models out of sample). The rate isn't always informative.

Common pitfalls

• Dollar-denominated debt under a peg. The peg's stability tempts borrowers to take cheap foreign-currency debt. When the peg breaks, debts double overnight in local terms — the "balance-sheet effect" that destroyed firms in the 1997 Asian crisis.
• Confusing intervention with policy. Buying or selling reserves to smooth volatility is not the same as committing to a target. Markets distinguish; policymakers sometimes don't.
• Underestimating exit costs. Once pegged, exiting is politically loaded — every devaluation is a defeat. Argentina spent a decade defending convertibility past its sell-by date.
• Treating "free float" as costless. Volatility imposes hedging costs on traded sectors. Small economies with big tradable shares often choose stability deliberately.