For decades, long-term investors have been told that broad benchmarks offer safety through diversification: if you own the market globally, you dilute single-stock risk, spreading exposure across countries, sectors and currencies. In theory, this provides better return assumptions and risk mitigation.

Yet in today’s more concentrated and dispersed markets, benchmark-level returns can conceal widening differences beneath the surface. Portfolios aligned to the same benchmark may appear similar, but behave very differently in practice. The distinction lies not in headline exposure, but in how risks are distributed and managed.

Why do the ‘Magnificent Seven’ still dominate?

The difficulty is that diversification assumes that no single group of companies dominates the outcome. Over the past decade, the S&P 500, once seen as a broad reflection of the US economy, has increasingly become dominated by technology and AI-related companies. In 2025, for example, the so-called “Magnificent Seven” tech stocks accounted for 42 per cent of the S&P 500’s 17.9 per cent return, while the information technology and communication services sectors together drove more than 63 per cent of the index’s overall gain. Excluding those sectors, the index would have returned just 6 per cent. While that’s not an argument that benchmarks are broken, it does highlight how headline returns can mask underlying concentration.

Industry experts have warned of the risks of becoming trapped by benchmarks – allowing them to dictate behaviour rather than serve as reference points. When an index becomes highly concentrated in a small number of stocks, investors who assume it represents maximum diversification may be taking more risk than they realise.

When diversification narrows

However, concentration is only one way in which benchmarks can change. Sector and country exposure have shifted too, says Habib Subjally, Senior Portfolio Manager and Head of Global Equities at RBC BlueBay, an investment management firm.

“Technology now represents roughly double the share of global indices than it did a decade ago, while the US accounts for around 72 per cent of the MSCI World index,” says Subjally. “People assume they’ve got maximum diversification and often they don’t.”

These shifts rarely happen overnight. As the biggest companies outperform, their weight in cap-weighted indices rises. Passive inflows then direct more money towards those same stocks, reinforcing their dominance. Gradually, a “neutral” portfolio can become increasingly concentrated.

The consequences of this become clearer when volatility returns and markets fall. Maria Satizabal, Portfolio Manager in Multi-Asset Credit and Asset Allocation at RBC BlueBay, points out that two portfolios tracking the same benchmark can deliver markedly different outcomes depending on design and risk management.

“Clients don’t experience benchmarks,” she says. “They experience drawdowns.”

These are the periods when portfolio values fall sharply and take time to recover.

She points to the stock and bond market decline of 2022 when two managers using the same US investment grade bond benchmark delivered very different results. One actively managed duration exposure, limiting losses to around 5 per cent during the sell-off, Satizabal says. The other focused more on credit risk and saw a drawdown closer to 14 per cent, while the benchmark fell about 16 per cent.

“Even though investors thought they were invested in something very safe,” she says, “the experience could be completely different because the shock came from rising rates rather than defaults.”

The hidden bias inside bond benchmarks

Part of the issue lies in benchmark construction. In bond markets, benchmark weights are determined by the amount of debt outstanding. In simple terms, the more a company or government has borrowed, the larger its presence in the index. “Those issuers with the highest weights are those with the most debt,” Satizabal says. “That doesn’t automatically mean they are the strongest credits.”

Unlike equity indices, which reward market success, bond indices give greater weight to the biggest borrowers. In benign conditions, that structure may go unnoticed. In periods of stress, however, heavier exposure to more indebted issuers can amplify losses.

That structural bias is only part of the picture. Other risks can be less visible until markets come under pressure. Liquidity can evaporate just when investors most need it – and bonds that look investment grade can be downgraded in weaker conditions. Benchmarks provide an anchor, says Satizabal, but “you really need to analyse drawdown, volatility in stress conditions and liquidity resilience.”

Cycles, not collapse

None of this means concentration is unprecedented. Ludovic Phalippou, Professor of Financial Economics at Oxford University’s Saïd Business School, cautions against treating current levels as historically unique. “Concentration today is clearly elevated,” he says, “but not historically unprecedented.”

The share of the largest 10 companies in major indices has fluctuated in cycles. In the late 1990s, mega-cap outperformance pushed concentration sharply higher before it fell in the 2000s. “Much of what we are seeing is a familiar return cycle playing out again,” says Phalippou.

That historical perspective should temper the temptation for alarmism. Concentration has risen before and then fallen back. But cycles can persist for years, and during those periods the risk profile of a benchmark can look very different to what long-term investors assume.

Cap-weighted indices are designed to reflect the market, but their structure can introduce unintended risks. Because the largest companies receive the greatest weight, strong performance from a handful of sectors can quickly reshape the index’s composition. In recent years, the rapid rise of mega-cap technology firms has increased sector concentration, leaving benchmarks more exposed to technology cycles, regulatory pressures and shifts in investor sentiment. As a result, indices that once appeared broadly diversified may now behave more like sector-tilted portfolios, reminding investors that market benchmarks evolve alongside the markets they track.

Intermediaries need to challenge the assumption that the benchmark is their maximum diversification proxy, Subjally argues. Cap-weighted indices reflect the balance of market participants – including hedge funds, high-frequency traders and short-term flows – as much as they reflect the needs of long-term savers.

Satizabal is careful to stress that benchmarks remain essential.

“Benchmarks are anchors,” she says. “They provide a reference point and a means of accountability.”

However, treating them as evolving tools rather than fixed truths may help to ensure that long-term objectives remain grounded in risk as well as return.