Brent vs XOM 5-day forward spread after Iran/Hormuz headline-intensity spikes
A tanker scare in the Strait of Hormuz should, in theory, bid a temporary supply premium into spot crude that an integrated equity like XOM treats as transitory. That's the thesis: Brent should lead XOM over the following week. It doesn't hold up.
Across three years of headline intensity, the 60 top-quintile spike days produced a Brent-minus-XOM five-day spread of -0.26%, versus -0.42% on the 544 quiet days. The gap runs the predicted direction, but only by 0.16 percentage points — a t-statistic of 0.18 and a p-value of 0.86, which is coin-flip territory. Brent beat XOM on just 42% of spike days.
The full breakdown, leg by leg, is below.
Over the past ~3 years, when headline intensity for Iran/Hormuz/tanker/sanctions keywords spikes into the top quintile of its prior 20-session distribution, does Brent crude's 5-day forward return outpace XOM's by more than on low-headline days? I expect Hormuz scares to embed a temporary supply premium in spot crude that equities discount as likely transitory, so Brent should lead XOM over the following week.
How this was measured
Built daily Brent prices from brent_daily_df and daily XOM closes from XOM minute bars, then kept their common trading days over the latest three-year window. Headline intensity on each day is the count of XOM_news titles/summaries containing Iran, Hormuz, tanker, or sanctions; non-trading-day articles are pushed to the next trading session. For each day, the 80th and 20th percentiles of the prior 20 sessions (current session excluded) define top-quintile spike and low-headline conditions. The outcome is the 5-trading-day forward return spread (Brent forward minus XOM forward), compared across conditions with Welch two-sample t-tests on the spread and on each leg separately.
The key numbers
Reading the numbers
On the 60 top-quintile headline days, Brent lagged XOM by 0.26% over the next five days versus a 0.42% lag on low-headline days, so the spike gap was only +0.16%. The Welch p-value of 0.857 means that difference is statistically indistinguishable from noise.
The charts
The box plot compares the Brent-minus-XOM five-day forward spread on top-quintile headline days against low-headline days. The spike group's mean is -0.26% (n=60) and the low-headline mean is -0.42% (n=544), so the spike box sits only a hair higher, by about 0.16 percentage points. But both distributions are very wide — spike spreads run from -13.1% to +25.8%, low-headline from -13.7% to +15.0% — so the two groups overlap heavily, which is why the p=0.857 says the gap is not a clear signal. For the question, this is the central test, and it does not show Brent decisively leading XOM after Iran/Hormuz/tanker/sanctions spikes.
The bars split the average five-day forward return into Brent and XOM under each headline condition. On top-quintile spike days, Brent averages -0.35% while XOM averages -0.08%, so Brent actually lags XOM. On low-headline days, Brent averages -0.10% and XOM +0.33%, a wider lag. The expected story was that Brent would lead XOM after supply-scare headlines, but the chart shows Brent underperforming XOM in both regimes; the only twist is that its underperformance is slightly smaller on spike days, matching the +0.16% spread gap rather than a true Brent lead.
Forward 5-day return summary
| Condition | N | Mean Brent 5d fwd | Mean XOM 5d fwd | Mean Brent−XOM spread | Frac spread > 0 |
|---|---|---|---|---|---|
| Top-quintile spike | 60 | -0.0035 | -0.0008 | -0.0026 | 0.4167 |
| Low-headline | 544 | -0.001 | 0.0033 | -0.0042 | 0.454 |
| All days | 694 | 0.0019 | 0.0041 | -0.0022 | 0.4481 |
The takeaway
No — Brent does not meaningfully outrun XOM over the five days after an Iran/Hormuz headline spike. If anything the raw numbers lean the opposite way for both legs: on spike days Brent averaged a 5-day return of -0.35% while XOM lost only -0.08%, and on quiet days XOM actually gained 0.33% against Brent's -0.10%. The spread you asked about (Brent minus XOM) was -0.26% after spikes versus -0.42% on low-headline days — a gap of just +0.16% in the direction you predicted, but so small that a t-statistic of 0.18 and a p-value of 0.86 put it firmly in coin-flip territory. That's about a 1-in-1.2 chance of seeing a difference this size from randomness alone. Reinforcing that: Brent beat XOM on only 42% of spike days and 45% of quiet days, so the typical outcome after a Hormuz scare is that the equity keeps pace or wins. The practical read is that any temporary supply premium that gets bid into spot crude isn't showing up over the following week — the market either prices it in immediately or fades it, and a 60-day event sample spread across three years isn't enough to distinguish a real effect from noise. This is a clean null, not a subtle one.
The fine print
- Headline intensity comes from XOM news coverage only, so Iran/Hormuz stories that never mention XOM are missed and the spike count is likely understated.
- After a headline burst, 5-day windows overlap heavily; the test treats them as independent, which flatters the apparent sample size.
- Rolling 20-session thresholds mean a single keyword headline can register as a 'spike' if the recent baseline was near zero.
- With only 60 spike days, the p-value is a rough guide rather than a precise probability.