Brent → Nifty, same day
0.132
statistically significant, but the sign is positive
Crude Oil × Nifty 50 Correlation Lab
India imports the large majority of its crude, so the oil-to-equity channel is one of the most repeated claims in Indian market commentary. This page tests it against every Nifty session from2006-01-02 to2026-09-23 —5,136 trading days — rather than restating it. The answer is not the one the folklore gives.
Beta (Nifty % per 1% crude)
0.062
a 10% crude move maps to roughly this % in Nifty
Variance in Nifty explained by crude
1.74%
R² of daily returns — the rest is everything else
Crude up → Nifty down
42.7%
below the 46.3% all-days base rate
After all controls
0.133
USD/INR + S&P 500 + VIX controlled (p = <0.001)
The short answer
The sign is wrong. Across 5,136 trading days, Nifty and crude are 0.132 POSITIVELY correlated — days crude rose, Nifty rose slightly more often than it fell (42.7% down). The familiar "crude up, Nifty down" rule does not hold in the daily data.
Put plainly: crude has a real but small statistical link to Nifty, and it is not the one the folklore claims. The whole-sample R² is 1.74% — crude moves explain about1.7% of the day-to-day variation in Nifty. The rest is rupee, global risk, sector earnings and everything else that moves the Indian market on its own.
The direction also fails the sanity check, and fails it in the opposite way to the folklore. On days crude rose, Nifty fell42.7% of the time — less often than the 46.3% base rate across all sessions — and averaged positive (+0.181%). If you followed the "crude up, Nifty down" reflex mechanically, this data says you would have been mildly on the wrong side of the trade.
What the data do support is the rupee leg of the story. Splitting crude-up days by what the rupee did: when USD/INR was strengthening (crude rising on global risk, not on an importer-specific supply shock), Nifty averaged +0.476%. When the rupee was weakening alongside crude — the genuine import-bill squeeze — Nifty averaged -0.139%. That gap is the oil-to-India channel, and it runs through the currency, not the barrel.
For contrast, Nifty's same-day correlation with the S&P 500 is0.363 and with USD/INR it is-0.379. Those are the relationships that actually explain Indian index moves. Oil is one input into them, not an independent driver of Nifty.
Crude and Nifty, side by side
Nifty 50 (right axis) against Brent and the derived rupee-crude (left axis), full sample. Notice how often the two move together rather than in opposition — the visual case against the naive rule.
Correlation matrix — every field against every field
Pearson correlation of daily % changes across Nifty open/high/low/close, Brent, and the derived rupee-crude. The Nifty-to-Nifty block is near 1 by construction; the interesting cells are the bottom row and column where crude meets Nifty.
Every day, plotted
Each point is one Nifty session: crude % change on the x-axis, Nifty % change on the y-axis, with the fitted regression line. A clearly negative slope would support the folklore; the shallow positive slope is the finding.
Does crude lead Nifty?
Correlation between crude today and Nifty k days later, for k = 0 to 10. The signal is concentrated at k = 0 and essentially gone by k = 2 — crude does not forecast Nifty beyond the current session.
What survives when you control for everything else?
Brent's correlation with Nifty as confounding factors are added one at a time. Because the raw link is already weak, the interesting result is that it barely changes: controlling for the rupee, global risk and volatility leaves crude's contribution essentially intact. That is a small but real effect — it is not explained away by the controls, and it is still far too small to trade on.
| Controls | Partial r | p-value | Std. beta | n |
|---|---|---|---|---|
| None (raw) | 0.132 | <0.001 | 0.132 | 5,135 |
| USDINR | 0.132 | <0.001 | 0.132 | 5,135 |
| SP500 + USDINR | 0.132 | <0.001 | 0.132 | 5,135 |
| VIX + SP500 + USDINR | 0.133 | <0.001 | 0.133 | 4,595 |
The partial correlation barely moves as controls are added, which means crude's small link is not simply a proxy for the rupee or global risk. For scale, Nifty's same-day correlation with the S&P 500 is0.363 and with USD/INR it is-0.379 — several times larger than crude's.
Is the relationship linear? (crude deciles → average Nifty return)
Days are split into ten equal groups by that day's crude % change. The average Nifty return (bars, truncated axis) and the share of down days (line) by group. A strongly negative relationship would show bars falling left-to-right; they do not.
Has the relationship changed over two decades?
Rolling 90-day correlation (and 252-day beta) of Nifty on crude. The line crosses zero and its level drifts between eras — the clearest evidence that there is no single stable 'crude number' for Nifty.
The relationship by era
The same-day correlation is not a constant — it is regime-dependent. It was clearly positive before the shale era, weakened through the 2010s, and turns negative from 2022. Any claim about "the" crude–Nifty correlation is really a claim about a particular period.
| Period | Same-day r | Beta | Avg Nifty on crude-up days | n |
|---|---|---|---|---|
| 2006-2009 | 0.224 | 0.183 | +0.409% | 985 |
| 2010-2013 | 0.258 | 0.187 | +0.225% | 1,000 |
| 2014-2017 | 0.130 | 0.050 | +0.134% | 987 |
| 2018-2021 | 0.141 | 0.042 | +0.171% | 991 |
| 2022-2026 | -0.060 | -0.019 | +0.001% | 1,172 |
The folklore, tested directly
Conditional outcomes: given a crude move, what did Nifty do on average, and how often did it fall? This is the table a retail trader actually acts on, so it is worth reading carefully — especially the rupee-split rows.
| Condition | Days | Avg Nifty | Nifty fell | Avg crude move |
|---|---|---|---|---|
| Brent upBrent daily return > 0 | 2,550 | +0.181% | 42.7% | +1.74% |
| Brent downBrent daily return < 0 | 2,371 | -0.104% | 51.4% | -1.76% |
| Brent up > 1%Brent daily return > +1% | 1,461 | +0.233% | 41.7% | +2.69% |
| Brent up > 2%Brent daily return > +2% | 752 | +0.328% | 38.8% | +3.86% |
| Brent up > 20%Brent daily return > +20% (the folklore rule) | 3 | -0.238% | 66.7% | +46.11% |
| Brent up + rupee upBrent up AND USD/INR up (oil cost rises twice) | 1,117 | -0.139% | 53.1% | +1.70% |
| Brent up + rupee downBrent up AND USD/INR down (rupee cushions the move) | 1,298 | +0.476% | 33.6% | +1.80% |
| Crude-INR up > 1%Brent x USD/INR daily return > +1% | 1,476 | +0.127% | 45.3% | +2.62% |
Note the extreme-move row: days crude rose more than 20% are so rare in the modern sample (3 days) that any "rule" about them is statistically meaningless. And the last rows show that when crude rises but the rupee is strengthening (crude's rise is being driven by global risk, not by an oil supply shock specific to importers), Nifty tends to be up.
Distribution of Nifty returns: crude-up vs crude-down days
Overlaid histograms of Nifty daily returns on days crude rose versus days it fell. If the 'crude up, Nifty down' rule held strongly, the crude-up distribution would sit clearly left of the crude-down distribution. The separation is slight.
Six real oil shocks, and what Nifty actually did
Each row is a dated, source-verified crude shock. The crude move and the Nifty response (same-day, and over the following 5 and 20 trading days) are measured from the underlying data, not asserted. The mix of signs — a fall in crude with a fall in Nifty (2020), a rise in crude with a flat Nifty (2022) — is the point: shock direction alone does not determine the equity response.
GFC oil peak — record $147/bbl
2008-07-11
Brent 136 → 144 USD/bbl (+5.8%). Nifty same-day -2.72%; next 5 days +1.1%; next 20 days +11.9%.
Crude averaged $90.7/bbl in January 2008 and touched a record $147 in July, then collapsed to roughly $32 by December. FIIs withdrew over $6bn and the Sensex lost more than 8,000 points on the year. Use as a confounded case: the Indian crash is overwhelmingly attributed to the global financial crisis, not to oil.
Sources: U.S. Energy Information Administration, The New York Times, Times of India
Libya civil war spike
2011-02-23
Brent 107 → 110 USD/bbl (+2.8%). Nifty same-day -0.58%; next 5 days +1.8%; next 20 days +1.6%.
Brent ran to a then-post-2008 high of $111.25 on 23 Feb 2011 and averaged $111.26 across 2011 — the first year the benchmark averaged above $100. Indian markets fell alongside global equities on 22-23 Feb. 2011 was Sensex's first annual decline in three years, though reporting attributes it to inflation, RBI tightening, a weaker rupee and the Euro crisis together rather than oil alone.
Sources: U.S. Energy Information Administration, CNBC
Shale/oil collapse — Brent drops 7% to $71
2014-11-28
Brent 77 → 72 USD/bbl (-7.1%). Nifty same-day +1.11%; next 5 days -0.6%; next 20 days -4.0%.
Brent fell 44% between June and December 2014, from roughly $112-115 to a four-year low of $71.25. Sensex and Nifty set record highs on the crude fall in late November 2014, and the RBI cut its FY15 current-account-deficit estimate to 1.3% of GDP. This is a supply-boom shock, so the naive 'crude up, Nifty down' rule predicts the wrong sign here - a useful check rather than a confirmation.
Sources: Wikipedia, Business Standard, Times of India
COVID demand collapse — Brent drops 22% to $35
2020-03-09
Brent 46 → 35 USD/bbl (-22.5%). Nifty same-day -4.90%; next 5 days -14.2%; next 20 days -12.8%.
Brent fell 22.5% in a single session on 9 March 2020 and about 65% across Q1 - the worst quarter on record. WTI famously settled at -$37.63 on 20 April 2020 as storage filled. Nifty fell roughly 29% from its December 2019 peak to its March 2020 trough. This is a demand shock rather than a supply shock, so crude and equities fell together; treat it as a contaminated observation rather than a clean oil signal.
Sources: U.S. Commodity Futures Trading Commission, U.S. Energy Information Administration, IIM Shillong
Russia invades Ukraine — Brent jumps 8% past $110
2022-03-02
Brent 103 → 119 USD/bbl (+15.4%). Nifty same-day -1.12%; next 5 days -1.6%; next 20 days +5.2%.
Brent rose about 25% in the week after the 24 February invasion and peaked at $139.13 in March 2022, the highest since 2008, before easing to $71 by June 2023. Sensex and Nifty each fell about 3% on 28 February 2022, and Nifty was down roughly 12% by June. Reporting puts the Indian correction at about four months. Our data show the largest single-day crude jump of the episode landing with Nifty essentially flat, then falling over the following week.
Sources: The Hindu BusinessLine, CNBCTV18
Hormuz closure — Brent's biggest one-day jump of the crisis, +12.5%
2026-03-12
Brent 91 → 102 USD/bbl (+12.5%). Nifty same-day -0.95%; next 5 days -2.7%; next 20 days +2.5%.
US-Israeli strikes began 28 February 2026 and the Strait of Hormuz was effectively closed. Brent rose from about $61 at the start of the year past $100 by mid-March, and the first quarter of 2026 saw the largest inflation-adjusted quarterly rise in data going back to 1988. India's import bill rose sharply, the rupee hit a record low, and FPIs sold off heavily in March. 12 March is the single largest daily Brent gain of the crisis, and Nifty fell on the day - the naive rule working as expected. But the price kept climbing to its peak in April, and Nifty rallied to new highs as crude made its high: a direct in-sample counterexample to treating the relationship as stable.
Sources: U.S. Energy Information Administration, International Energy Agency, Business Standard, The Hindu BusinessLine, CNBC
Caveats & statistical notes
- Correlation is not causation. This page measures co-movement, not a causal effect of oil on Indian equities.
- Daily financial returns are serially correlated, so the p-values shown are optimistic; treat significance as a rough guide, not formal inference.
- Partial correlation is a linear control. Crude may still matter through nonlinear channels (level-dependent inflation, second-round effects) that a linear residual does not capture.
- The derived rupee-crude series (Brent × USD/INR) is a model, not a traded price, and omits local basis, freight, taxes and MCX premium/discount.
- "Crude up 20% → Nifty down" days are so rare in the post-2006 sample that the folklore rule is based on a handful of observations, mostly from crises that also hit equities for unrelated reasons.
Data sources
Market data: Crude Oil Prices: Brent - Europe (DCOILBRENTEU), Crude Oil Prices: West Texas Intermediate (DCOILWTICO), Indian Rupee to U.S. Dollar Spot Exchange Rate (DEXINUS), Historical index data - Nifty 50 and India VIX, S&P 500 daily index history. Nifty 50 and India VIX daily history from the National Stock Exchange of India; Brent, WTI and USD/INR from FRED; S&P 500 daily history from Yahoo Finance. All series were retrieved on 2026-09-27.
Macro and event citations are linked inline throughout the event study above. The oil-to-inflation mechanism discussed on this page is documented by the Reserve Bank of India; see RBI Bulletin (Jul 2025) (a 10% crude rise adds ~20bp to Indian inflation) and Mint Street Memo No. 17 ($10/bbl adds ~43bp to the CAD/GDP ratio). These establish that oil does matter for India's macro variables — which is exactly why it is worth asking whether it matters for the index, and the data say largely it does not.
Educational analysis only. Not investment advice.