XOM minus SPY quarterly excess return: accelerating vs decelerating real GDP growth quarters
The hypothesis looks clean on paper: faster GDP growth should mean more oil demand, higher crude prices, and a tailwind for XOM. The data from the past three years tells a different story — in accelerating-growth quarters, XOM trailed SPY by an average of 12.4% and only outperformed once in five tries. In decelerating quarters, it beat the market three times and averaged a 12.5% excess return.
That inversion is striking, but the sample is thin: ten quarters split evenly between regimes, with a p-value of 0.053. It is a suggestive lean, not a proven reversal. The full breakdown below walks through the methodology, the quarterly excess returns, and why one strong quarter could still flip this picture.
For XOM over the past ~3 years, does its quarterly return beat SPY's quarterly return more often when real GDP growth accelerates quarter-over-quarter? I expect XOM to outperform during economic acceleration because rising GDP fuels higher energy demand and oil prices, lifting the supermajor.
How this was measured
Daily close prices of XOM and SPY resampled to calendar-quarter end, quarterly returns computed, and XOM minus SPY excess return derived. Real GDP quarterly data from real_gdp_quarterly_df used to calculate quarter-over-quarter growth rate (pct_change). Each quarter classified as accelerating if current growth exceeds prior quarter growth, else decelerating. Welch two-sample t-test compares XOM-SPY excess return distributions across the two regimes. Sample limited to roughly 12 quarters due to 36-month history window.
The key numbers
Reading the numbers
On average, XOM trailed SPY by 12.45% in accelerating quarters and beat it by 12.48% in decelerating quarters. The difference isn't statistically clear (p=0.053), so the data don't support the idea that economic acceleration helps XOM.
The charts
The left bar sits below zero at -12.45%, while the right bar is above zero at +12.48% — a mirror image of what your hypothesis predicted. Instead of XOM beating SPY when GDP growth accelerates, the average excess return was negative in accelerating quarters and positive in decelerating quarters. The two bars are nearly equal in size but point in opposite directions, so at the mean level the pattern is the reverse of what you expected.
Look at where the two distributions sit: the accelerating group is centered near -12.45% and drops as low as -32.58%, while the decelerating group is centered near +12.48% and reaches a strong +44.45% quarter. Even though both groups have only five quarters each, the accelerating quarters are mostly losses relative to SPY and the decelerating quarters are mostly wins. This matches the headline fractions — XOM beat SPY in only 20% of accelerating quarters but 60% of decelerating quarters — so the visual spread reinforces the conclusion that acceleration did not help XOM.
Quarterly XOM vs SPY excess return and GDP acceleration status
| Quarter | XOM-SPY excess | GDP growth | Accelerating |
|---|---|---|---|
| 2024Q1 | 0.0675 | -0.0342 | No |
| 2024Q2 | -0.0455 | 0.0322 | Yes |
| 2024Q3 | -0.0304 | 0.0095 | No |
| 2024Q4 | -0.0996 | 0.0191 | Yes |
| 2025Q1 | 0.1643 | -0.0368 | No |
| 2025Q2 | -0.1983 | 0.0289 | Yes |
| 2025Q3 | -0.0217 | 0.0135 | No |
| 2025Q4 | 0.0467 | 0.0159 | Yes |
| 2026Q1 | 0.4445 | -0.0369 | No |
| 2026Q2 | -0.3258 | 0.0285 | Yes |
The takeaway
The short answer is no — over the past three years XOM did not beat SPY more often when GDP growth accelerated; if anything, the evidence tilts the opposite way. In the five accelerating-growth quarters, XOM lagged SPY in four of them, averaging a −12.4% excess return, while in the five decelerating quarters it beat SPY in three and averaged +12.5%. The t-test gives p ≈ 0.053, meaning about a 5-in-100 chance this gap is pure luck — just shy of the conventional 0.05 threshold. But with only 10 quarters, 5 per bucket, this is a suggestive lean rather than a proven reversal: one strong quarter could flip the picture. Practical takeaway: the simple “faster GDP growth → XOM outperforms” story did not hold in this window, so don’t trade that hypothesis with much conviction; the signal is weak and small-sample.
The fine print
- Only 10 quarters total (5 per regime) — a single quarter can move the averages substantially.
- Acceleration is just growth rising vs the prior quarter; it ignores magnitude and whether growth is above trend.
- GDP is released with a lag; same-quarter alignment measures association, not a tradeable leading signal.
- Oil-specific shocks (OPEC, geopolitics) likely matter more than GDP direction and are not isolated here.