Concentration risk: when diversification hides shared exposures

More holdings do not always mean more diversification

Capitalization-weighted indices are designed to represent markets, not to keep every risk source equal. When a small number of very large companies outperform, their index weights rise and a diversified benchmark can become increasingly dependent on them. S&P Dow Jones Indices reported that the ten largest companies represented almost 40% of the S&P 500 by mid-2025, a level not seen since the mid-1960s.

Concentration is not automatically a defect. Market leaders can possess strong balance sheets, durable franchises, and superior growth. The risk is that investors may not realize how much of the portfolio’s outcome now depends on the same earnings assumptions, valuation regime, technology cycle, or geopolitical exposure.

Security count measures breadth. It does not measure independence between the risks those securities carry.

Concentration has several layers

• Name concentration: a few issuers account for a large share of portfolio value or expected return.
• Sector and factor concentration: different companies respond to the same technology cycle, interest-rate regime, valuation factor, or investor narrative.
• Geographic concentration: holdings depend on the same end market, manufacturing base, commodity route, or regulatory jurisdiction.
• Supply-chain concentration: apparently unrelated businesses rely on the same semiconductor foundry, cloud platform, logistics corridor, mineral processor, or internal-service location.
• Policy concentration: many holdings are vulnerable to the same tariff, export-control, sanctions, tax, or industrial-policy decision.

Measure the overlap, not just the weights

Top-five and top-ten weights are useful first checks. The Herfindahl-Hirschman Index and the related ‘effective number of holdings’ provide a more complete view of position concentration. But portfolio analysis should also aggregate company-level exposures: revenue by country, production and sourcing locations, critical dependencies, and sensitivity to the same regulatory events.

This look-through approach can reveal why two funds with different names—or two companies in different sectors—may still share the same underlying risk. It also makes stress tests more realistic: instead of shocking one ticker at a time, the portfolio can be tested against a disruption that affects every company using the same supply route or market.

Build a deliberate concentration budget

• Decide which concentrations are intentional sources of conviction and which are accidental by-products of benchmark design.
• Set monitoring thresholds for top holdings, sectors, factors, countries, and common operating dependencies.
• Compare replacement candidates on both fundamental quality and hidden geopolitical exposure.
• Rebalance when a position’s weight or shared exposure exceeds the portfolio’s risk budget—not merely because it has performed well.

The objective is not equal weight for its own sake. It is clarity: knowing where the portfolio is making a concentrated bet, why that bet exists, and what could cause several holdings to fail together.