Section 3

Macro comparison

Equity against gold and silver, the interest-rate channel, and crude / USD-INR sensitivity — each with the statistical caveat attached where it belongs, not buried in a footnote.

The single most important caveat on this page

Every macro correlation here uses one observation per financial year — n = 10. A correlation needs |r| > 0.63 to be significant at the 5% level with ten points. Not one rate correlation in this study clears that bar. They are all statistically indistinguishable from zero and should be read as hypotheses, not findings. Where a relationship is economically sensible and above the bar, it is called out explicitly below.

Equity vs gold vs silver

Annual return by financial year (INR)

Gold and silver are MCX continuous futures — gold quoted ₹ per 10g, silver ₹ per kg.

Growth of ₹100

Compounded across the ten financial years, price only. Gold spent six straight years trailing equities before dominating FY25–FY26.

Risk-on or risk-off?

Classified by whether equities beat gold. "Risk-off" means gold outperformed — it does not mean equities fell.

FY	Nifty 500	Gold	Silver	Equity − gold	Regime

Equities won 6 of 10 years — but look at which four

The risk-off years are FY20 (COVID), FY23, FY25 and FY26 — the two crisis periods plus the recent stretch. Gold did nothing at all in FY17 (−0.3%) and trailed equities in six of the seven years FY18–FY24. Gold behaved as a regime-dependent hedge, not a steady one. FY26 was extreme: silver +140.7%.

The interest-rate channel

Correlation between the year's repo-rate change and each sector's return, across the ten years. The shaded band is inside ±0.63 — the region where a correlation is statistically indistinguishable from zero at n=10.

corr(repo change, sector median return)

Positive = the sector did relatively better in years when the repo was hiked.

pale = inside the noise band (|r| ≤ 0.63) solid = clears the significance bar No sector clears it.

The textbook rule is contradicted twice, badly

FY20: the repo was cut 185bp — and rate-sensitive sectors still lost to defensives by 20.6pp. The binding constraint was asset quality and a credit freeze, not the cost of funds.
FY23: the repo was hiked 250bp — and rate-sensitives won by 13.7pp, during a credit-growth and NIM-expansion upcycle.
Across all ten years the correlation between the repo change and Financial Services' return was +0.17: financials did better in hiking years. The credit cycle dominates the policy rate.

Crude and USD/INR

Method note — why these use excess returns

Correlating raw sector returns with crude gives a spurious near-universal positive relationship (Metals & Mining +0.89) purely because FY21 had crude +161% and equities +76%. Subtracting the index isolates the differential sensitivity. Even then, USD/INR results are confounded: the rupee weakened most in FY20 and FY26, both weak-equity years, so "rupee weak" is entangled with "bad equity regime".

Sector	corr crude (raw)	corr crude (excess)	corr USD/INR (raw)	corr USD/INR (excess)

The one macro result worth taking seriously

FMCG shows −0.83 excess correlation with crude — the only relationship here that is both economically sensible (crude is an input and logistics cost for packaged consumer goods) and above the |r| > 0.63 significance bar. FMCG's relative performance suffered in the crude-spike years FY21, FY22 and FY26.