Section 2

Sector rotation

Which of the 20 broad industries led each financial year, how leadership rotated, and whether any sector's strength actually persisted.

Survivorship bias applies to every figure on this page

Sector medians are computed from the 2026 constituent list, so sectors that lost members to delisting or demerger over the decade are flattered. Sector sample sizes vary widely (n=3 to n=52) — small-n sectors move for idiosyncratic reasons.

Median return heatmap

Rows sorted by average rank across the decade. Cells show that sector's median stock return for the year. Darker green = stronger, darker red = weaker.

Weaker Stronger ±60% saturation · hover a cell for the sample size
How to read this

Very few rows are consistently one colour. Metals & Mining and Capital Goods show the two clearest multi-year themes (commodity reflation, and the domestic capex/PSU cycle), while Information Technology swings from the best sector in FY18 and FY21 to the worst in FY23 and FY26. Rotation is the norm, not the exception.

Rank rotation

Rank 1 = best median return that year, out of sectors with n≥3.

Did the leader keep leading?

Each row: where the previous year's best sector ranked the next year (blue), vs where the previous year's worst sector ranked (violet).

Previous year's leader vs laggard — next-year rank
Lower is better (1 = top sector). If leadership persisted, blue dots would cluster at the left and violet at the right.
Sector leadership persists about as often as it reverses

The leader continued in FY18→FY19 (IT), FY20→FY21 (Chemicals), FY21→FY22 (Metals) and FY23→FY24 (Capital Goods) — four cases. It reversed hard in FY19→FY20, FY22→FY23 and FY24→FY25 — three cases. Notably, laggards also bounced: IT went from FY17 laggard to FY18 leader. There is no mechanical "buy last year's winning sector" trade here.

Rate-sensitive vs defensive

Rate-sensitive = Financial Services, Realty, Auto, Construction. Defensive = FMCG, Healthcare, Telecom, Power. Each row shows both medians for that year.

Median return: rate-sensitive vs defensive, by FY
The right-hand number is the spread in percentage points (defensive − rate-sensitive). The repo change is shown in the table below.
FYRate-sensitiveDefensiveSpreadRepo changeConsistent with "rates down ⇒ rate-sensitives win"?
The popular rule mostly does not hold

The rule holds in roughly four clean cases and fails in two spectacular ones. Across the ten years the correlation between the repo change and Financial Services' return was +0.17 — financials did better in hiking years. In FY20 the repo was cut 185bp and rate-sensitives still lost by 20.6pp (asset quality, not cost of funds). In FY23 the repo was hiked 250bp and rate-sensitives won by 13.7pp (a credit-growth and NIM-expansion upcycle).

Concentration of winners

HHI of the top 30 by sector. 20 equal sectors would give 0.05; 1.0 is total concentration.

Herfindahl index of the top 30 performers, by FY
Dashed line = 0.05 benchmark for 20 equal sectors. Winners clustered into 10–11 of 20 sectors, never all 20.