Ask "what determines the exchange rate?" and the honest first answer is the same as for any price: supply and demand for the currency at the margin, driven by everyone who needs to convert. Exporters bringing dollars home, importers buying dollars, tourists, and, dwarfing all of them in the modern world, investors moving capital. Over 90% of daily FX turnover is financial, not trade-related; the current account is a trickle next to the capital flood. So to know what moves a currency you must know what moves capital, and that comes down to five forces, each dominant over a different horizon.
How do interest-rate differentials move a currency?
Force 1: interest-rate differentials (the dominant short-run driver). Money chases yield. If Indian government bonds yield 6.5% and comparable US bonds yield 4.5%, capital wants to flow into rupee assets to earn the extra 2%, and that demand for rupees strengthens the rupee — in the short run. This is why a surprise rate hike usually lifts a currency and a surprise cut usually sinks it, and why the FX market hangs on every central-bank word (M7.02, §4.8). But there is a deep catch, and it is the hinge of everything that follows: the extra yield cannot be a free lunch, or everyone would borrow dollars and buy rupee bonds without limit. The no-arbitrage condition that resolves this (covered interest parity) says the high-yield currency must be expected to depreciate by roughly the yield gap, so the level of rates matters less than surprises and changes in the expected rate path. Rate differentials dominate day-to-day and quarter-to-quarter FX.
Advertisement
What does purchasing-power parity say?
Force 2: purchasing-power parity (PPP), the long-run gravitational anchor that fails short-run. The law of one price says identical goods should cost the same everywhere once expressed in a common currency (otherwise you could arbitrage: buy cheap, ship, sell dear). Extended to a whole basket, absolute PPP says the exchange rate should equalize price levels, so if a basket costs $100 in the US and ₹6,000 in India, the PPP rate is ₹60/$. The Economist's Big Mac Index is the famous, deliberately jokey version: compare the local-currency price of a Big Mac across countries to see which currencies look cheap or dear versus the dollar. The key operational cousin is relative PPP, which is what an analyst actually uses:
Expected % change in the exchange rate (home per foreign) ~ home inflation - foreign inflation
That is: the currency of the higher-inflation country should depreciate over time by roughly the inflation gap. If India runs 5% inflation and the US 2%, the rupee should lose about 3% a year against the dollar over the long run, and, remarkably, it roughly has, averaging ~3–3.5%/yr depreciation over decades (currency risk makes this the centerpiece for a cross-border investor). PPP fails badly in the short run. Currencies deviate from PPP for years, even a decade, and for real reasons: many goods and services (haircuts, housing, healthcare) are non-traded and can't be arbitraged; there is the Balassa–Samuelson effect (poorer, faster-growing economies have systematically lower price levels, so their currencies look "undervalued" on PPP but that gap is structural, not an arbitrage opportunity); and capital flows swamp goods flows for years at a time. So use PPP the way you use a long-run mean: a gravitational anchor that says where a currency should drift over 5–10 years, not a prediction of next quarter. A currency far below PPP is cheap and faces long-run upward pull; a currency far above is dear — but "far from PPP" can persist long enough to ruin anyone who trades it impatiently.
How do terms of trade move a currency?
Force 3: terms of trade (the price of what you sell versus what you buy). A country's terms of trade is the ratio of its export prices to its import prices. When the price of what you export rises relative to what you import, each unit of exports buys more imports. The country is richer, its trade balance improves, and its currency tends to strengthen. The cleanest examples are commodities. When oil prices spike, an oil exporter (Saudi Arabia, Russia, Norway) enjoys a terms-of-trade windfall and its currency firms; an oil importer (India, which buys ~85% of its crude abroad) suffers a terms-of-trade shock: its import bill balloons, its trade deficit widens, and the rupee comes under pressure. This is why "oil up" is, all else equal, "rupee down," a linkage you can practically trade, and one that makes oil India's single largest external vulnerability (see commodities as macro signals). Terms of trade dominate for commodity-heavy economies and over the medium term.
Why do safe-haven flows move currencies?
Force 4: safe-haven flows (fear, over hours and days). In a global panic, capital does not chase yield. It chases safety, and it does so violently and fast. A handful of currencies are safe havens: the US dollar above all (for reasons the eurodollar system makes mechanical: the world owes dollars, so in a crisis everyone scrambles for dollars), the Japanese yen (a creditor nation whose investors repatriate in stress), and the Swiss franc. When the VIX spikes, these three tend to strengthen regardless of their interest rates, while risk-sensitive and emerging-market currencies (the rupee, the rand, the real, the lira) sell off. Safe-haven demand explains the otherwise-baffling fact that the dollar often rises during a US-centered crisis (2008): the crisis increases the world's desperation for the very dollars it is short. Fear dominates FX over hours and days and overrides every other force while it lasts.
How does growth move a currency, and which force wins?
Force 5: relative growth and productivity (the slow, deep current). Over long horizons, capital is drawn to economies that grow faster and more productively. Higher expected returns on real investment pull in FDI and equity flows, strengthening the currency, and faster productivity growth can support a stronger real exchange rate (the Balassa–Samuelson mechanism again, in reverse: a catching-up economy's real rate tends to appreciate). This is the deep case for the rupee being less weak than the pure inflation differential implies when India's growth and productivity are strong, and it is why structural-reform news (GST, IBC, manufacturing incentives) can matter for the currency over years. Growth and productivity dominate the multi-year trend beneath the noise.
The synthesis an analyst carries: fear (hours–days) → rate surprises (weeks–quarters) → terms of trade (quarters–years) → PPP and growth/productivity (5–10 years). Different force, different horizon. Most bad currency calls come from applying a long-run force (PPP: "the rupee is cheap") to a short-run question ("what will it do this quarter?"), or vice versa.