The Fed and Indian equities: four channels, and why they do not simply add

What the Fed actually sets, and what it does not

The US Federal Reserve sets one thing: a target range for the overnight rate at which American banks lend each other reserves. That one setting reaches an Indian share price through four separate chains, each running at its own speed, and any two of them can point in opposite directions on the same day.

Which is why the Fed cut, so Indian markets should rise is not so much wrong as unfinished. It compresses four mechanisms into one arrow and then wonders why the arrow keeps failing. The four are worth separating because they act on different things: one is arithmetic, one is capital allocation, one is sentiment and leverage, and one is about what the Reserve Bank of India can comfortably do next.

Start with the boundary of the Fed's actual power, because most of the confusion lives here. The Fed sets an overnight rate. It does not set the ten-year yield. A ten-year yield is conventionally read as the market's expectation of the average overnight rate across the next ten years, plus the extra compensation lenders want for tying money up that long. Ten years is about 3,650 days, so a cut today changes one of them. If the market simultaneously decides that the average of all those days is higher than it thought, the long yield rises on the day of a cut. That is not a market misunderstanding the Fed. It is the long yield doing what it is.

Everything below traces pressure through a channel. None of it forecasts a market, because none of it can: earnings, domestic flows, crude and Indian policy are all moving at the same time, and any one of them can more than offset a rate effect.

Channel one: the discount rate, and the part the Fed does not control

The first chain is pure arithmetic and needs no investor to do anything. A share is a claim on money a company will produce far into the future, and converting future money into today's money means dividing by a rate. Move the rate, and the answer moves before a single share has traded.

That mechanism, including why a long-duration growth company reprices harder than a steady cash generator on the identical rate move, is set out in full in US yields and Indian share prices. This article assumes it rather than repeating it.

What belongs here is the link from the policy rate to that yield, and it is looser than the shorthand suggests. The Fed moves the front of the curve directly. The rest of the curve is the market's expectation of the path, and the market forms that expectation from everything it hears, not from the decision alone. A cut delivered alongside language that shrinks the expected total number of cuts can raise the two-year yield the same afternoon. The policy rate went down; the rate that discounts Indian cash flows went up.

So the honest statement of channel one is narrow. A change in expected US policy moves the global rate anchor, and a change in the anchor changes present values by arithmetic. A change in the policy rate itself moves the anchor only to the extent it changed what people expected — which is nothing at all when the decision lands where the curve already had it, and can be the opposite of the decision when the language cuts across it. The section below on reading the surprise rather than the setting is about exactly that gap.

Channel two: the dollar, the rupee, and the hurdle for foreign capital

The second chain runs through capital, and it takes weeks rather than milliseconds.

Tighter expected US policy tends to firm the dollar, because holding dollars pays more relative to the alternatives, and the dollar is the currency the world borrows in. A firmer dollar tightens global financial conditions in a way that lands hardest on markets funded by borrowed dollars. The mechanism, and why the dollar index is the single broadest summary of that condition, belongs to the dollar index and emerging markets.

Two things happen to an Indian portfolio at the same time, and they push the same way, which is what makes this channel self-reinforcing while it runs.

The hurdle rises. When safe dollar assets pay more, the return a global allocator demands before holding an Indian asset rises with it — not because India got worse, but because the alternative got better. Whatever no longer clears the hurdle becomes a candidate to be sold.

The currency eats the return. For an unhedged dollar investor, the realised return is the Indian return plus the currency move. A weakening rupee subtracts from the dollar outcome, so the Indian return required to compensate goes up too. The rupee's own transmission chain — imported inflation, the current account, why sudden moves matter more than slow drift — is covered separately.

One qualification that changes how you read the data. A flow number is a footprint, not a force. Net foreign portfolio flow records transactions that have already happened. The mechanism is the hurdle rate; the flow figure is evidence about it, published afterwards. Treating the flow series as the cause reverses the arrow — and it is the commonest reading error on the whole page.

The other qualification is who else is buying. A market funded largely by foreign capital transmits this channel almost directly; one with a large steady domestic bid can absorb foreign selling without the price clearing where it otherwise would. Which of the two describes India in any given stretch is a question about the flow record rather than about theory — the daily foreign and domestic cash figures are published, they are their own subject, and this article attaches no number to them.

Channel three: risk appetite, which responds to the reason, not the rate

The third chain is the fastest and the least mechanical. Indian equities carry global beta — on a broad risk-off day they mostly fall with everything else, largely independent of anything Indian. What that day looks like across the tiles is its own article.

Here is the part specific to the Fed, and it is where the crude version breaks hardest. Channel three responds to why the Fed acted, not to what it did. A cut delivered because inflation came down is a different event from an identically sized cut delivered because employment is deteriorating fast, even though the rate ends in the same place and channel one is indifferent between them.

The transmission is through leverage and the price of protection rather than through a discount rate. Cheaper funding makes leveraged positions cheaper to carry, which supports the appetite for risk; a policy move read as a response to something breaking raises the demand for downside protection instead. Those show up on different tiles — equity volatility, bond volatility, credit spreads, the yen cross — and they can contradict the yield tile in the same session. Credit is worth watching first, because spreads tend to register funding stress before equity indices do, and volatility itself splits into a US gauge and a domestic one that do not measure the same thing.

There is also a plumbing route with no sentiment in it at all. A great deal of global risk-taking is financed in the currency that is cheapest to borrow, and a fast move in that funding currency forces positions to be closed regardless of anyone's view on India. The yen carry trade is the standing example, and it is why the Macro page handles the yen cross twice over — once as an ordinary scored tile, and once as a stress trigger that sits outside the score entirely. Why it needs both is the subject of the last section.

Channel four: what the gap does to the RBI's room

The fourth chain is the one the crude version misses entirely, and it is the slowest of the four. It does not act on prices at all. It acts on the set of decisions the Reserve Bank of India can take without a currency consequence it would rather avoid.

Two claims here, and they sit in different tiers, so it is worth being exact about which is which.

The mechanical one. The difference between Indian and US rates is the price of hedging the rupee. A forward exchange rate is not anybody's forecast of the currency — it is pinned by arbitrage against the two interest rates, because borrowing in one currency, converting, lending in the other and selling the proceeds forward has to earn nothing. So the rate gap is the forward premium: widen it and a rupee hedge costs a foreign investor more, narrow it and it costs less. That is arithmetic, not opinion, and it holds whatever anybody intends. An investor who declines to hedge has not escaped the term; they have chosen to carry it as currency risk instead.

The judgement one. A wider gap is conventionally read as giving the RBI more comfortable room to cut without adding to pressure on the rupee, and a narrow gap as less. That is a reasonable reading that experienced people contest, and it is not a rule. The RBI has published no target differential and describes policy as set for domestic inflation and growth. The policy rate is decided in Mumbai, on Indian data, and it can move less than, more than, or opposite to the US rate over any given stretch.

What makes this channel worth tracking anyway is the trade-off buried in it, which nothing on any dashboard displays. Defending a currency and supporting domestic growth pull against each other: the stance that keeps the differential comfortable is the tighter one, and tighter costs domestic activity. So channel four does not resolve into a direction. It resolves into a constraint — a set of options that has become wider or narrower — and constraints do not show up on a tile because they are not a number.

Note also the asymmetry in speed. Channel one reprices in a session. Channel four plays out over policy meetings, which is to say months. Reading them off the same screen on the same afternoon invites you to treat a fast mechanism and a slow one as though they were commenting on the same thing.

Why the four do not simply add

Because they arrive at four different speeds and any of them can point against the rest. Set side by side, the reason the one-arrow version fails is structural rather than unlucky.

Now the case that makes the point concrete. Take one cut of a given size, delivered for two different reasons, and trace the same four chains. The cells below say which way each chain pushes — not what any market did, which is a different question and not one this article answers.

Identical policy action. Channels one and three end up on opposite sides in the right column, and channel two stops having a clean sign at all. The Fed's action is not the signal. Its reason is. Anyone summing four arrows into a market call is adding quantities that are not on the same scale, arriving at different speeds, some of which are not directions in the first place.

The trade-off in reading it this way is honest and worth stating: you give up the single clean answer. Four channels with different speeds and conditional signs will not collapse into a view, and no amount of staring will make them. What you get instead is the ability to say which chain a given day is running through, which is a question with an answer.

The setting is not the news: read the distance from what was priced

Here is the specific mistake, named so it is recognisable: reading the setting instead of the surprise. A reader sees that the Fed cut, and reasons forward from the cut. But the cut was not new information to the market that prices Indian assets.

The mechanism is straightforward once stated. Short-dated yields already embed the market's expectation of the policy path — that is most of what a two-year yield is. If a decision lands exactly where the curve had it, nothing was learned, and the arithmetic that channel one runs on has no new input to run on. What moves is the gap between what happened and what was already priced, plus whatever the accompanying language did to the expected path beyond this one meeting.

Two consequences follow, and both are checkable rather than arguable.

A decision day with no market reaction is not a market ignoring the Fed. It is what an expectation being met looks like. Going looking for an explanation of the non-move is looking for a cause of nothing.

The interesting tile on a policy day is whichever one had it wrong. If the short-dated yield moves and the long end does not, the surprise was about the next few meetings. If the long end moves more, the surprise was about the path or about the compensation lenders want for holding duration — and if that repricing is large enough to change the shape of the curve itself, that has its own reading. If neither moves much but volatility and credit do, the surprise was about the reason rather than the rate, which is channel three.

None of that predicts anything. It tells you which chain the day ran through, which is the only question a macro dashboard is equipped to answer.

Reading this on the Macro page

Everything above is readable off public data and a calculator. FNOTrader's Macro page assembles the pieces in one place, and several things about how it presents the Fed are worth stating plainly — starting with the most useful fact in this section, which is a negative.

There is no Fed input in the composite score. The page carries a central-banks card showing the current policy or reference rate for the RBI, the Fed, the ECB and the Bank of Japan. None of them feeds the score. The composite is built from the market tiles only, plus net foreign flow folded in at scoring time — so what the page scores is not the Fed's decision but the prices that decision is transmitted through: the US two-year and ten-year yields, the dollar index, volatility, credit spreads, the yen cross. That design is the whole argument of this article rendered as a data structure. You cannot score a policy rate against Indian equities directly, because the rate reaches them only through the channels above, and the channels have their own prices.

The stance label is computed from the rate series, not from what officials said. The card compares the latest reading with the value about three months earlier and labels the result: hiking if the rate rose by more than 0.02 percentage points, cutting if it fell by more than that, on hold otherwise. So on hold means the rate itself has not moved by more than a rounding margin over roughly a quarter. It says nothing about guidance, minutes, or the expected path — the very things that, as the section above sets out, do most of the work. Read the label together with the three-month change beside it, and read both as history rather than as a stance in the rhetorical sense.

One more piece of small print on that card: the RBI row is not the repo rate. It uses the India call money rate — the overnight rate banks actually transact at, which is the RBI's own operating target and which it steers towards the repo — as an honest proxy, because the public series the card is built from carries no direct live repo rate. The US row is the fed funds target upper bound, the top of the range described at the start of this article. Neither is a mis-statement, but a monthly proxy for an observed market rate and a daily administered target are not the same kind of number, and a reader comparing the two closely should know which is which.

Colour means impact on Indian equities, not the direction of the number. This inverts several tiles relative to instinct. A falling dollar index shows green, because a softer dollar loosens global financial conditions. A rising USD/JPY also shows green, because a weaker yen means the carry trade is intact rather than unwinding. The colour answers what this does to us, never did this go up.

The weights are our judgement, and the divisor is not fixed. The score is a weighted average of the tile contributions, score = 100 × Σ(w_ic_i) ÷ Σ(w_i), with the dollar index carrying the largest weight at 0.20, the US ten-year next at 0.15, then Brent at 0.12, the yen cross and US volatility at 0.10 each, and the US two-year among several inputs at 0.06. Net foreign flow is added at scoring time with a weight of 0.12, saturating at ₹5,000 crore of net cash-segment flow in either direction; it is not one of the tiles. Because a failed feed drops out of the numerator and the denominator both, the divisor is the sum of the weights that actually reported that day, not a constant. Regime boundaries sit at ±20, with a stress override at −35 or on a one-day US volatility jump of 20% or more or a one-day USD/JPY fall of 1.2% or more. Every one of those numbers is FNOTrader's modelling judgement about what matters to Indian equities — a considered view, not a measured constant and not a fitted coefficient. A different reasonable weighting produces a different score from identical inputs.

The saturation built into those contributions is what the doubled-up yen handling is for, and it is the clearest thing on the page about what a weighted score cannot express. Each tile's contribution is its daily move divided by a saturation point and then clipped to the range −1 to +1. For the yen cross that saturation point is a 1% move, so a 1% fall and a 5% fall contribute exactly the same thing: the full −1 a tile is allowed, carrying its weight of 0.10. A clipped average cannot express a disorderly move — the clipping is what stops one violent tile from swamping every other input, and the price of it is that the extreme and the merely large arrive at the score looking identical. The separate stress trigger on a one-day USD/JPY fall of 1.2% or more exists to say the thing the clipped average has no way of saying. The same logic sits behind the volatility trigger. Any composite you meet, ours included, is worth checking for this: ask where it saturates, and what it therefore cannot tell you.

Finally, the correlation heatmap will not help you with the Fed, and it is better to know that in advance. It computes Pearson correlation of Nifty daily returns against a fixed list of ten market drivers over a 30, 60 or 90-day window. There is no policy-rate entry in that list — nor, worth knowing before you go looking for it, India VIX. The nearest things to the Fed in it are the US ten-year yield and the dollar index, which is to say channels one and two and nothing else. And a correlation that flips sign between two windows is describing the windows, not discovering a relationship. The wider set of tiles, and how to read them together rather than one at a time, is the subject of the macro pillar.

Common questions

Continue reading

More in Macro & Markets · App: Options Analytics · Definitions: glossary · Free tools: calculators · All: every article