Bernstein: Indian equity holdings diverge: domestic institutions add exposure as crowded positioning points to sector rotation

How has the ownership structure of Indian equities changed?

On a market-capitalization-weighted basis, domestic institutions held 18.3% of BSE 500 stocks as of September 2026, up from 18.0% at the end of the second quarter; foreign institutions’ share fell from 16.3% to 16.2%. Domestic institutions first overtook foreign investors at the end of 2025, and the gap widened further in the third quarter.
Foreign investors reduced their ownership of large-cap stocks in the third quarter, from 18.4% to 17.8%; their ownership of mid-caps rose from 13.5% to 14.3%. Domestic institutions added exposure to large-, mid- and small-cap stocks, with their small-cap ownership rising from 14.5% to 15.2%.
Retail investors’ direct ownership of small-cap stocks fell from 14.1% to 13.2%, but remained above foreign investors’ 10.3% and mutual funds’ 12.2%.

In the third quarter, foreign ownership of the technology sector rose from 14% to 15%, while holdings in other sectors were broadly flat or edged lower. Foreign ownership was below its respective ten-year average in every sector except healthcare, industrials and energy excluding Reliance.
Domestic institutions increased their holdings in most sectors, with ownership of financials, communications and technology rising to 22%, 20% and 20.1%, respectively; healthcare and real estate edged lower.

Bernstein compares total current institutional holdings in individual stocks with their averages over the past 1, 3 or 5 years. Since 2006, the group of stocks with the largest increases in holdings relative to history has generally performed weakly over the following 3 to 12 months, underperforming the market by around 2.8 to 3.1 percentage points annualized over the long term.
The underweight screen also relies on historical comparisons: after excluding the 20% of stocks with the lowest deviations in holdings, the eightieth-percentile portfolio generated historical annualized excess returns of 0.8% and 3.1% relative to the 3-year and 5-year averages, respectively.
The comparison window changes the results: as of 26 October 2026, the underweight portfolio screened against a 5-year benchmark had outperformed the market by 1.5 percentage points year to date, while the crowded portfolio had underperformed by 1.7 percentage points; under the 3-year benchmark, the two groups had underperformed by 2.3 percentage points and outperformed by 7.6 percentage points, respectively.

Each quarter, Bernstein goes long the three least crowded sectors and short the three most crowded. Since 2009, the long-short strategy has returned 4.5% annualized; the long sectors returned 17.0% annualized, compared with 13.7% for the market over the same period.
As of 26 October, the long-short strategy had gained 2.8% year to date, while the market had fallen 9.7% over the same period; however, the strategy fell 0.9% when the market rebounded in the second quarter. The firm believes positioning mean-reversion signals are more effective when the market is under pressure, while a recovery in risk appetite may favor crowded sectors.
Based on the latest sector rankings, Bernstein’s rotation view favors long positions in energy, automobiles, consumer discretionary and real estate, while avoiding insurance, telecommunications, materials and mining, and industrial products. The implications of positioning differ across sectors: historically, increases in foreign holdings of technology stocks have been positively correlated with performance in the following quarter, while increased holdings by domestic institutions in banks and telecommunications have also tended to exhibit momentum.