The era of 30 trillion: Quantitative private equity shifts from a race to elimination competition.

2026-05-24 08:14

Zhitongcaijing
The scale of private equity has crossed a critical threshold. Leading institutions are expanding through scale, devouring their own excesses, while the price-volume factors continue to retreat and have not yet been repaired. Where does the excess come from, and why does the retreat occur, are becoming more crucial selection criteria than historical rankings.
The scale of quantitative private equity has passed a critical point. Estimates show that by the end of 2025, the scale of quantitative funds will reach 3.22 trillion yuan, accounting for over 43% in the private equity securities investment funds; in 2020, this scale was only 0.76 trillion yuan. The fastest expanding strategy is equity quantification: index enhancement, stock neutrality, and quantitative stock selection together account for over 80%.
The Zhejiang Merchant Securities Financial Engineering Team, including Chen Aolin, pointed out in a report on May 21st, "As the scale reaches 3 trillion, the current situation has shifted from a race to an elimination game, and stable excess returns or the manager's core competitiveness come into play." The focus of this statement is not on the scale itself, but on the changes after the scale: excess returns are no longer easy to obtain, and the gap between managers is rapidly widening.
The most striking data comes from differentiation. By the first half of 2026, the estimated calculation indicates that the difference in excess returns between the 90th percentile and the 10th percentile managers of the top 500 index incremental products has widened to around 20%; the difference between the 70th percentile and the 30th percentile is also close to 8%. More crucially, the median is around 0, implying that a considerable number of products have not outperformed the price index.
Drawdowns have also become more difficult to explain. Since 2024, the fluctuation amplitude and frequency of excess returns in index incremental products have increased. In the drawdown in August 2025, factors such as low volatility, momentum, small market capitalization, and liquidity have shown clear drag, while growth factors have contributed positively to excess. Using one or two style labels to summarize performance differences in the past is no longer sufficient.
The larger the scale, the scarcer the excess returns.
From 2020 to 2025, the total scale of private equity securities investment funds has risen from 4.3 trillion yuan to 7.45 trillion yuan. Quantitative funds are the main source of growth, with the scale increasing from 0.76 trillion yuan to 3.22 trillion yuan; non-quantitative funds have also increased from 3.54 trillion yuan to 4.23 trillion yuan, experiencing several years of consecutive contraction in between.
Within the realm of quantification, index enhancement remains the largest strategy, with a scale of approximately 1.19 trillion yuan by the end of 2025, accounting for nearly 40% of the total scale of quantitative strategies. Stock neutrality is around 0.74 trillion yuan, and quantitative stock selection is approximately 0.54 trillion yuan. Quantitative CTA is 0.37 trillion yuan, while quantitative arbitrage and quantitative options have scales of 0.11 trillion yuan and 0.03 trillion yuan, respectively.
After the scale expansion, the pressure on excess returns begins to show.
From the beginning of 2019 to February 2026, the cumulative excess returns of the top three broad-based index incremental strategies have continued to rise: approximately 141% for the top 1000 index incremental, 109% for the top 500 index incremental, and 94% for the top 300 index incremental. However, after 2024, the volatility has significantly increased: the top 300 index incremental had negative excess returns in 2024, the top 500 index incremental has been in continuous drawdown since July 2025, and the slope of excess returns for the top 1000 index incremental has slowed down.
This change in trend is basically synchronized with the pace of scale expansion. From 2020 to 2021, the scale of quantification doubled from 0.76 trillion yuan to 1.46 trillion yuan, and the excess returns of the top three index incrementals all fell, with the excess return of the top 1000 dropping from 18.2% to 6.6%. From 2022 to 2023, the scale remained relatively stable, and excess returns were partially restored. By 2024, as the scale approached 2 trillion yuan, the excess returns of the top 300 index incremental turned negative, while the excess returns of the top 500 and 1000 also dropped to a low point. In 2025, the scale jumped to 3.22 trillion yuan, and the excess returns of the top three index incrementals rebounded, but this was largely overshadowed by the extreme market performance of small and medium-sized caps in the first half of the year.
The advantages of top managers will also be swallowed up by scale.
Scale is the enemy of excess returns, and this applies equally to top managers.
Take H Quant as an example: during the period of performance explosion from 2019 to 2021, its relative excess returns compared to the average of the same strategy once reached a peak of about 30%. However, as the management scale increased, the excess gradually narrowed after 2022, basically returning to the industry average.
The path of K Investment is similar. During the rapid expansion of scale from 2023 to 2024, there was still a significant excess advantage, but after 2025, the relative advantage began to decline and gradually approached the industry average; by 2026, there were even cases of slightly underperforming the industry average.
This means that the logic of relying on top-tier labels for selection in the past is becoming ineffective. The larger the scale, the more difficult it is to bypass issues like strategy capacity, crowded trades, and factor decay. It's one thing for managers to generate excess returns, but it's another to maintain those excess returns after expansion.
Differentiation is not only apparent at the top and bottom, but the middle layer is also beginning to widen.
In the estimated data for the first half of 2026, what is most noteworthy is not just the gap between the strongest and weakest performers, but also the gap that is starting to emerge in the middle layer.
In the top 500 index incremental products, the difference in excess returns between the 90th percentile and the 10th percentile is approximately 20%, which is close to or even greater than the extreme level of the same period in 2024. The gap between the 70th percentile and the 30th percentile is about 8%, indicating that the difference is not only evident in a few tail-end products, but is spreading to a wider range.
The median is around 0, which means that the cost of choosing the wrong product has increased. In the past, the core selling point of index enhancement products was stable excess returns; now, this premise is no longer firm. Relying solely on historical annual excess returns may misinterpret style premiums as management capabilities; solely looking at short-term rankings can lead to buying into drawdowns after style switches.
The more effective question becomes: where does the excess come from? When drawdowns occur, is it due to factor failure or overly concentrated portfolio exposure? If it's the latter, the speed of recovery usually doesn't just depend on market rebounds.
Alpha sources are narrowing, and price-volume factors are no longer tailwinds.
Behind the performance differentiation lies a change in the structure of Alpha factors.
Until before 2026, there were generally at least three categories among the four major classes of factors - fundamental factors, analyst expectations, price-volume factors, and high-frequency low-frequency factors - that could consistently provide positive excess. When multiple factors were simultaneously effective, managers with different methodologies could find sources of returns, and the performance gap naturally would not be too large.
In the past year, this landscape has changed. The main factors that still maintain positive excess are fundamental factors and analyst expectations; price-volume factors and high-frequency low-frequency factors have continued to draw down since August 2025, with no trend of recovery.
The center of excess returns has also clearly shifted downwards. Before the third quarter of 2025, the annualized excess center for the CSI 500 and CSI 1000 index incrementals was approximately 10% and 13% respectively; since the third quarter of 2025, the CSI 500 has dropped to about 0.75%, and the CSI 1000 has dropped to about 7.5%.The increase is only about 4.5%.This will directly amplify the differences between fund managers. Combinations with a higher proportion of fundamental factors and a more balanced factor structure recover from drawdowns relatively faster. An example is Manager Y: fundamental factors account for over 30%, while factor accounts for about 50%, and this round of excess drawdown has recovered to around 0.7%. In comparison, a combination represented by Manager K has a real-time factor number reaching tens of thousands, with over 90% being factor factors. When the factor factors experienced drawdown after August 2025, the excess drawdown amplitude was higher than the market average and has not completed recovery.
The main reason for drawdowns is often not due to factors performing poorly.
To explain the increased drawdown, looking solely at factor fluctuations is not enough. A more intuitive approach is to divide the stock pool by market capitalization and residual volatility: large-cap, mid-cap, small-cap, high-volatility, low-volatility, to form six style domains. This can break down excess drawdown into two parts: deviations of the stock pool and excess factors within the stock pool.
Taking the CSI 500 enhancement as an example, during the drawdown from August 8 to September 30, 2025, the excess drawdown in the CSI 500 enhancement range was -4.13%. During the same period, low-volatility factors saw a drawdown of -8.32%, momentum factors -6.95%, small-cap factors -7.22%, and liquidity factors -5.33%; while growth factors showed a +2.60%.
This indicates that not all factors fail simultaneously during drawdowns. The directional differences of different factors are significant, and the composition of the combination determines how much impact the net value ultimately bears.
The breakdown of liquidity factors can better illustrate the issue. From August 8 to September 1, 2025, the related portfolio drawdown was -3.75%, of which stock pool deviations contributed -3.11%. The total internal deviation within the CSI 500 contributed -3.77%, while allocation to other stock pools contribute +0.66%. Looking at the internal excess factors of the stock pool, the CSI 500 recorded -0.70% internally, while other stock pools recorded -0.14%.
In other words, this drawdown was primarily not because liquidity factors severely failed in all stock domains, but because deviations from the base in the portfolio dominated the losses, especially the deviations within the base.
The next round of screening for quantitative private equity will have stricter standards
After 30 trillion, quantitative private equity will continue to be an important force in private equity funds, but the industry's winning and losing strategies have changed.
The core issue in the past was whether excess returns could be achieved; now two more questions need to be asked: whether excess returns rely on a single style and whether they can be maintained after scale expansion; when drawdowns occur, whether they are short-term factor disturbances or the composition structure exposed to unsustainable deviations.
This is also the meaning of the elimination match. Not all managers will disappear, but under the same index enhancement label, product curves may be completely different. Stability of excess returns, risk control capabilities, and balance of factor structure will be more important than historical rankings.
A boundary needs to be maintained: the above calculations of returns and drawdowns rely on specific test intervals and samples and do not represent future results. If future market data distribution is inconsistent with history, the model may also fail. This article does not constitute a recommendation for fund products or involve recommendations for any holding stocks.
This article is reproduced from "Wall Street See", GMTEight Editor: Jiang Yuanhua.