Because the five pairs trade at wildly different absolute levels — roughly ₹95 to the dollar versus ₹1.15 to the rouble — comparing them on a single chart only works if every series is indexed to the same starting point. The chart below rebases all five to 100 at their first available trading day (15 August 2016) and tracks the rupee's cost of buying one unit of each currency since.

Four of the five lines move almost in lockstep, rising from 100 to roughly 145–148 over the decade — the rupee buying about 45–48% more of the dollar, dirham, riyal or yuan today than it did in August 2016. The rouble is the outlier, and not in the direction the other four suggest: it sits at just 104, the shallowest ten-year rise of the five, but it got there by way of the sharpest single move on the chart — the vertical plunge and rebound around 2022, when the rouble first crashed against every currency in the wake of the Ukraine invasion sanctions, then recovered hard as Russia's central bank imposed capital controls and its energy-export earnings kept flowing in roubles.

All changes are in INR terms — a positive number means the rupee has weakened against that currency (more rupees needed to buy one unit) over the stated period. Source: derived daily series, as above.

Dollar — the reference case. USD/INR has depreciated in an almost textbook straight line: +48% over ten years, or roughly 4% a year compounded, tracking the structural gap between US and Indian inflation and interest rates rather than any single shock. It is also, by construction, the currency every other pair in this piece is measured against, since none of the other four trade INR directly.

Yuan — the steepest recent slide. CNY/INR shows the fastest 1-year and 3-year rupee depreciation of the five (+15.5% and +24.0%), even though its 10-year move (+44.8%) is slightly gentler than the dollar's. That gap between the recent and long-run numbers is the signal: China's own currency management — the yuan's managed depreciation through 2023–25 as its economy slowed, then a partial 2026 stabilisation — has pushed more of the rupee's fall against the yuan into the last three years than into the whole prior decade. Section 6 of the "What India Is Actually Paying For" companion piece already flagged China's own near-zero domestic inflation (0.5%) as an outlier among India's suppliers; this chart shows that low inflation has not stopped the rupee from falling faster against the yuan than against the dollar recently.

Rouble — the one pair that broke the pattern. RUB/INR's ten-year change (+4.1%) is the smallest by a wide margin, and its three-year change (+32.0%, the largest of the five) is a mechanical artefact of measuring from a post-crash low rather than genuine three-year weakness. The honest read is that the rouble spent 2022 far outside its normal range in both directions — a currency briefly worth less than half of its pre-war rate, then briefly worth 50% more than that low, before settling into a level not dramatically different from where it started. India's rupee-rouble trade infrastructure (Section 3 of "Can Rupee Trade Actually Offset...") exists precisely because this kind of volatility, not the ten-year trend, is what actually complicates rouble-denominated trade.

Riyal and dirham — the dollar in two other names. SAR/INR and AED/INR move almost exactly with USD/INR at every horizon in the table — unsurprising, since both currencies are hard-pegged to the dollar (SAR at 3.75, AED at 3.6725, both fixed for decades). Their ten-year, five-year, three-year and one-year changes sit within a few tenths of a percentage point of the dollar's in every column. There is no independent "riyal trend" or "dirham trend" to read here beyond the dollar's own.

Trend and volatility are different questions. A currency can end a decade roughly where it started (the rouble) while being, day to day, the riskiest currency on the list by a wide margin — and that is exactly what the data shows.

RUB/INR's annualised volatility (52.2%) is roughly seven times the dollar's and eight times the dirham's — a currency that trades in a fundamentally different risk regime, not just a more volatile version of the same one. USD, AED and SAR cluster tightly at 6–10%, with SAR's slightly higher reading (9.6% against AED's 6.0%, despite both being hard dollar pegs) most likely reflecting noisier historical data around the Saudi peg rather than a real difference in risk — the dirham's series tracks its 3.6725 peg almost perfectly, while the riyal's shows more scattered ticks around 3.75. CNY sits a notch above the pegged currencies at 7.0%, consistent with a managed float that is looser than a hard peg but far tighter than a freely traded currency.

Every volatility figure above measures INR against one specific counter-currency. A separate, useful comparison is how the rupee's overall volatility ranks against other emerging-market currencies as a class — and on that measure, India's own numbers are unusually calm. Per the RBI's own October 2024 Monetary Policy Report, the rupee has been the least volatile among major emerging-market currencies in recent periods. The Economic Survey 2023-24 put a number on it: INR's coefficient of variation was 0.58 in FY2023-24, its lowest in recent years, despite the same period carrying global geopolitical risk, rising interest rates and volatile commodity prices that hit EM currencies broadly.

Every number for CNY/INR, RUB/INR and SAR/INR in this piece is a derived cross-rate — INR/USD divided by the USD price of each currency — because no direct, officially published INR reference rate exists for any of the three. FBIL (Financial Benchmarks India Pvt Ltd, the RBI-recognised publisher of India's official daily reference rates since July 2018) publishes direct rates against only six currencies: USD, GBP, EUR, JPY, AED and IDR. The yuan, rouble and riyal are absent.

That gap is not a data-collection inconvenience; it is itself part of the story this series has been tracking. It means every rupee-yuan, rupee-rouble or rupee-riyal transaction in India's real economy is already, mechanically, a two-leg conversion through the dollar — exactly the kind of friction the Special Rupee Vostro Account framework and the India-UAE Local Currency Settlement System (Section 2 of "Can Rupee Trade Actually Offset...") exist to route around. The UAE is the one relationship among these five where a direct settlement mechanism and a direct FBIL reference rate both already exist; China, Russia and Saudi Arabia have neither.

Every rate in this piece ultimately traces back to a single daily number: the official USD/INR reference rate. Where that number comes from, and how it has changed, is its own story — and it is a much more mechanical, and much more recently reformed, process than "the market decides" suggests.

The rupee has not always floated. From 1947 to 1971 it was pegged to the pound sterling; after Bretton Woods collapsed it briefly stayed pegged to sterling, then in 1975 India moved to a basket peg — the rupee's value fixed by the RBI within a ±5% band around a weighted basket of major trading-partner currencies, initially 14 currencies, later narrowed to five. That basket-peg mechanism is the direct conceptual ancestor of today's NEER/REER index below: the difference is that in 1975–92 the basket set a policy target, and today it only produces a reporting index. The transition to today's regime came via the 1992–93 Liberalised Exchange Rate Management System (LERMS, a transitional dual exchange rate), unified into a single, market-determined rate in March 1993 — still the regime in force.

Who actually calculates today's number, and how. From 1993 to 2018, the Foreign Exchange Dealers' Association of India (FEDAI) set the daily USD/INR spot fixing by polling at least five empanelled banks between 11:40am and noon, applying a ±3-standard-deviation rule to discard outlier quotes, and averaging what survived. On 10 July 2018, that responsibility passed to FBIL (Financial Benchmarks India Pvt Ltd — jointly formed by FIMMDA, FEDAI and the Indian Banks' Association, incorporated December 2014), which switched the method from polling bank quotes to a transaction-based calculation: real trade data pulled from CCIL's and Refinitiv's electronic platforms, a randomly selected 15-minute window inside 11:30am–12:30pm, a minimum of 10 trades totalling at least $25 million to qualify, the same ±3σ outlier rule applied to the surviving trades, then averaged. The rate is published around 1:30pm every business day. The shift from opinion-based polling to observable-transaction data mirrors the post-LIBOR-scandal reform wave that reshaped benchmark-setting globally — a CCIL working paper on the redesign explicitly frames the old method as vulnerable to the kind of manipulation that discredited panel-poll benchmarks elsewhere.

The dollar-as-pivot method this piece uses for CNY, RUB and SAR is not an improvisation — it is how the RBI itself computes effective exchange rates. The central bank's published methodology for its Nominal and Real Effective Exchange Rate indices (NEER and REER) derives most bilateral INR cross-rates the same way: crossing the official USD/INR reference rate against the ruling EUR/USD, GBP/USD, USD/JPY and other USD-quoted rates, since India does not publish direct reference rates for most of its trading partners' currencies either.

RBI's version of this exercise is far larger than the five pairs in this piece: the current NEER/REER series (rebased to 2015-16 = 100 in a 2020 revision) covers a 40-currency basket, weighted by each country's share of India's total trade, together accounting for roughly 88% of India's merchandise trade. NEER is the trade-weighted average of the rupee's bilateral rates against that basket; REER adjusts NEER for relative inflation using the CPI as deflator, so a rise in REER signals the rupee has become more expensive in real, inflation-adjusted terms against its trading partners as a group — the single-currency version of the "is India getting more or less competitive" question this piece asks pair by pair.

Trade balance figures from DGCI&S TradeStat, FY2025-26 (as reported in "Can Rupee Trade Actually Offset..."). Currency and volatility figures as derived above.

The pattern that falls out of this table is not the one a simple "which currency is riskiest" framing would predict. India's single largest deficit (China, −$112.1bn) sits against a currency with no direct settlement route and a recently steepening slide — a combination that means both the settlement friction and the exchange-rate cost are getting worse at the same time, with no mechanism in place to offset either. India's most volatile currency exposure by far (Russia) sits against a deficit less than half the size of China's, but is the one relationship where a real, if imperfect, rupee-settlement mechanism already exists — precisely because the volatility and the sanctions friction made dollar settlement itself the harder problem to solve first. And the two hard-pegged currencies, Saudi riyal and UAE dirham, offer the least FX news of all: whatever India pays or saves against them is really a story about the dollar, mediated through a fixed exchange rate that has not moved by more than a rounding error in a decade.

A companion piece on this site ("Can Rupee Trade Actually Offset...") has already tracked the mechanics of India shifting more of its trade into direct rupee settlement — Special Rupee Vostro Accounts, the UAE's Local Currency Settlement System, Russia's rupee-surplus workaround. Reading that alongside the currency data in this piece produces a caveat worth stating explicitly: growth in rupee-denominated trade would change some of what's shown above, but not most of it, and not evenly across the five currencies.

The part that could move the rupee's own value, not just settlement friction. "Pumping" more rupees into global trade — pushing more of India's invoicing into INR — has a direction-dependent effect on the currency itself that the settlement-mechanics framing above doesn't fully capture, and it runs opposite ways depending on which side of the trade balance a partner sits on. Rupee-invoicing India's exports (the USA relationship, Section 7's one surplus) creates genuine foreign demand for rupees — a buyer abroad has to acquire INR to pay an Indian exporter, which is currency-supportive in the same direction a stronger current account already is. Rupee-invoicing India's imports from deficit partners (China, Russia, the UAE, Saudi Arabia — all four in Section 7) does the opposite: it hands foreign exporters rupees they have no natural use for, since they buy far less from India than India buys from them. Russia's accumulated rupee surplus (Section 3 of the companion piece) is the live example of exactly this mechanism, not a Russia-specific quirk — scale rupee invoicing up across all four deficit relationships and the same oversupply-of-unwanted-rupees dynamic would recur at each one, each needing its own RBI workaround (the treasury-bill absorption used for Russia) to avoid becoming latent sell pressure on the rupee whenever a holder tries to convert the balance back. Net effect on USD/INR itself is genuinely ambiguous without knowing the invoicing mix in advance: it is not simply "more rupee trade strengthens the rupee," and for India's four deficit relationships in this piece, the structural pressure plausibly runs toward the RBI needing to keep absorbing offshore rupee balances rather than toward organic appreciation.