Currency Regimes: Who Decides What a Currency Is Worth

A pegged currency can look remarkably calm for years and then move 30% in a morning. That is not a market anomaly; it is what the end of a policy looks like.

The spectrum, from free float to no currency at all

Regimes form a continuum rather than two categories, and the labels matter less than what a government has actually committed to. A free float lets the market set the rate with no target and little or no intervention. Rates move continuously, sometimes substantially, and that visible volatility is often mistaken for instability when it is in fact the mechanism by which shocks are absorbed. A managed float — sometimes called a dirty float — has no announced target but involves intervention to smooth movement or resist a direction. This is the most common arrangement among large economies and the least transparent, since the market must infer the authorities' tolerance rather than read it. A crawling peg adjusts a target rate on a pre-announced schedule, often to accommodate higher domestic inflation, which is an attempt to have both stability and adjustment. A band permits movement within stated limits, with intervention at the edges — so the rate is market-determined in the middle and policy-determined at the boundaries. A hard peg fixes the rate at a declared level and defends it. A currency board goes further: the domestic currency is fully backed by reserves of the anchor currency under a legal commitment, which is more credible than a policy peg precisely because it is harder to abandon. And dollarisation or monetary union is the endpoint — a country uses another currency, or shares one, giving up an independent exchange rate entirely. Monetary union has its own history, covered in the macro pillars, and the Gold Standard was an early global instance of the same trade-off. What every regime other than a free float has in common is a commitment, and every commitment has a cost that someone bears.

The impossible trinity, and why pegs break

One framework organises this whole subject. A country can have at most two of three things: a fixed exchange rate, free movement of capital, and an independent monetary policy. Not one of the three is objectionable; the constraint is that all three together are not available. Fix the rate and allow capital to move freely, and interest rates must follow whatever the anchor country does — you have surrendered monetary policy. Keep policy independence with a fixed rate, and capital movement must be restricted. Keep policy independence and open capital markets, and the rate must float. Every regime on the spectrum is a choice about which of the three to give up, and reading a country's regime tells you what it has chosen to sacrifice. That framing is worth more to a reader than any list of regime types. Now why pegs break, which is the practical part. Defending a peg means selling foreign reserves to buy your own currency when it is under pressure, and reserves are finite while pressure need not be. Four mechanisms follow. Reserve exhaustion: a sustained deficit or capital flight drains the reserves that make the commitment credible, and the market can observe the drain. Divergent fundamentals: if domestic inflation persistently exceeds the anchor country's, the real exchange rate appreciates even with the nominal rate fixed, making exports progressively uncompetitive — pressure builds from the arithmetic rather than from speculation. The interest-rate cost: defending a peg often requires raising rates sharply, which damages the domestic economy, so the defence has a political price that rises with its duration. And self-fulfilling attack: if participants believe a peg will break, selling the currency makes breaking it more likely, and the cost of defending rises. The result is a distinctive risk profile. A pegged currency exhibits very low measured volatility right up until the break, and then a single enormous move. Historical volatility therefore understates risk in exactly the regimes where it is lowest — which is one of the more important observations in this pillar, because low measured volatility is routinely presented as evidence of safety and here it is closer to evidence of a suppressed adjustment. Two things this portal will not do. It will not assess whether any particular peg will hold, because that is a forecast about a policy decision and this pillar has already established what it thinks of forecasts. And it takes no position on whether pegs are good policy — the arguments for stability and credibility are real, the arguments about surrendered flexibility are real, and the debate is genuinely live among economists.