-1.png)
The Robeco NextGen Global Small-Cap Equity UCITS ETF has just gained its London listing. The product is active, machine learning-driven, and aimed squarely at the part of the market that has spent years being too inefficient for passive products to cover well and too labour-intensive for most active managers to cover cheaply. Whether a quantitative algorithm can thread that needle is the interesting question.
First, some context. Robeco entered the European active ETF market in October 2024 and promptly won New ETF Issuer of the Year at the ETF Stream awards. By the end of 2025, the firm had crossed one billion euros in AUM across its ETF range. That is a genuinely impressive start. The Dutch asset manager has been building out its product shelf methodically, covering quantitative global equity, thematic and now small-cap, alongside a growing fixed income suite. The NextGen Global Small-Cap ETF is one of the more conceptually interesting additions to that range.
The strategy uses what Robeco calls its “NextGen” quantitative model. This is machine learning applied to the small-cap universe, sifting through a broad opportunity set of developed market companies and ranking them against a set of return factors including intrinsic value, quality, price momentum, analyst sentiment, and low volatility. The model is not replacing human judgment with a black box so much as systematically processing a volume of data that no conventional fundamental analyst team could cover at the same scale across global smaller companies.
That last point matters for understanding why small-cap specifically makes sense for this approach. The large-cap universe is extremely well-covered. Every major bank has analysts who have spent careers modelling Apple, Microsoft, and their equivalents. The informational edge available to any participant in that space is thin. Small-cap developed market equities are a different story. Coverage is patchy, data is noisier, and mispricings that a disciplined systematic model can identify persist longer because fewer people are looking for them. This is where quantitative factor investing has always had its strongest theoretical case, and where machine learning techniques that can ingest alternative data sources alongside traditional fundamentals have genuine potential to add value.
The ETF wrapper matters here too. Traditional small-cap active funds have often had to choose between deep coverage and cost efficiency. An active ETF structure provides the transparency, daily liquidity, and relatively low cost overhead of an index product while retaining the stock selection discipline of an active process. For European investors, getting access to a systematic small-cap strategy through a UCITS ETF listed on the London Stock Exchange is a cleaner, cheaper proposition than the alternatives that existed before.
The LSE listing adds to existing exchange coverage across the continent, broadening the product's accessible investor base to include UK wealth managers, advisers, and institutional allocators operating within UCITS-friendly mandates. Given that small-cap has underperformed large-cap for a sustained period, particularly outside the US, there is also a contrarian case for considering the segment at a moment when flows and attention are concentrated elsewhere. Whether Robeco's machine sees what the market currently does not is, of course, the question that only time will answer.
What can be said now is that the product fills a meaningful gap in the European active ETF landscape. Global small-cap systematic strategies in ETF format are not common. The combination of Robeco's quantitative heritage, a machine learning stock selection process, and an accessible ETF structure is at least a coherent and original proposition in a market that does not always offer either of those things.
Comments