Systematic Edges: The APB Enhanced Indexing Forum

Systematic Edges: The APB Enhanced Indexing Forum

Asian Private Banker (APB) recently co-hosted a lively and thought-provoking roundtable discussion alongside its partner, Aberdeen Investments, at the Fullerton Hotel in Singapore on 13 May 2026.

The Enhanced Indexing Forum brought together Nick Millington, the firm’s head of systematic index solutions, and the heads of discretionary portfolio management (DPM) from seven of Asia’s leading private banks.

The forum focused on the evolution of enhanced index investing. It explored how these strategies aim to deliver incremental returns within a disciplined, risk-controlled framework for APAC private wealth clients.

Millington opened with Aberdeen’s broader capabilities. As at the end of May 2026 the firm’s Quantitative Investment Solutions (QIS) team manages approximately US$170 billion across four core strategic pillars: enhanced indexing, ESG strategies, traditional indexing (across both equities and fixed income), and customised solutions.

“We have navigated a wide array of market fluctuations over the years, and throughout that time, we’ve grown our business by listening to our clients,” Millington noted.

At its core, enhanced indexing aims to deliver the benchmark market return plus a consistent layer of alpha in a highly disciplined, risk-controlled framework.

“We are beta one. We have constraints to make sure we’re diversified. We don’t want our returns to come from any one huge stock position or sector or region,” Millington explained.

While the firm targets to outperform standard market indices using factors, Millington emphasised: “the portfolio is king”.

“I can talk about factors, individual stocks, and sectors, but the end investor ultimately buys and experiences a unified portfolio. Our experience as fund managers has taught us that we need to continually improve our process through time, not just finding the best sources of return through the factors, but making sure that the portfolio contains all of that insight,” Millington said.

Against a backdrop of ongoing fee compression, the discussion underscored how enhanced indexing can serve both as a complementary building block and as an alternative to traditional passive Exchange-Traded Funds (ETFs) and active strategies.

“Everyone is looking for value for money, but they are also looking to build exciting propositions. Some of the more exciting assets that you can put into portfolios can be a bit more expensive, whether it’s private assets or thematics. Having a core that delivers value for money can help you expand the proposition into a variety of new and interesting assets,” he said.

Millington observed that clients frequently utilise these strategies as foundational building blocks for core equity allocations, but they also serve as a complement to existing active managers.

”We firmly believe if you find different alphas, it’s a good idea to combine them because you get a better risk-adjusted return. The consistent approach is key, because if our customers want to buy more than one of our strategies, maybe from different regions, they already understand how they can fit together,” Millington noted.

The roundtable featured a rigorous, transparent debate. The DPM heads asked highly technical questions to parse the nitty-gritty of the strategy, diving deep into the decision-making process, risk management, factor combination, performance during market drawdowns, and tracking error management.

They also shared the current state of quantitative investing within their own private banking platforms, highlighting where enhanced indexing could add the greatest value for clients.

The discussion was detailed, candid and technically rigorous and gave participants a clearer view of how Aberdeen’s enhanced indexing approach is built, managed and positioned within client portfolios.

Here are the edited highlights of the discussion.

Participants:

Harsh Agarwal, Managing Director, Head of DPM, Asia Pacific, Deutsche Bank Private Bank
Yves Bogni, Head Mandate Investment Team, UBS Global Wealth Management
Robin Chay, Executive Director, DBS Bank
Daniel Furer, Head, Discretionary Portfolio Management, Standard Chartered
Paras Gupta, Head of Investment Services, Southeast Asia and Head of Discretionary Portfolio Management, Asia, Union Bancaire Privée
Arjun Panchapagesan, Head of Portfolio Management Asia, CEO EFG Asset Management Singapore, Hong Kong
Qian Su, Head of Investment Management, Asia, Indosuez Wealth Management

How do you differentiate between Smart Beta, Multi-Factor ETFs, and Enhanced Indexing?

Nick Millington, Aberdeen: Some of you might remember—or perhaps still invest in—what the industry called Smart Beta. When it first launched, it used style factors but often paid less attention to the standard benchmarks we are all measured against. That meant they could outperform, but often with greater variation versus market-cap indices. Enhanced indexing starts from the benchmark, then systematically tilts the portfolio to add factor exposures, while tightly controlling active risk.

Plus, even if a Smart Beta fund uses multiple factors, its risk budget may end up biased toward just one of them. What we try to do with enhanced Indexing is to maintain a balanced exposure across our three core factors (momentum, quality and value). While they outperform over the long run, they are cyclical in the short run. Maintaining exposure to all three simultaneously gives you the true benefit of diversification and smooths out the noise.

That sounds simple, but it’s actually quite hard to execute. That’s why we’ve built our own proprietary risk models to use in our portfolio optimiser. Why is that so important? Because if you buy an off-the-shelf vendor risk model, it won’t have your unique definitions of risk or your specific factor equations. Because the vendor model doesn’t recognise your signals, it may end up underestimating your true risk.

Five years ago, we started building our own region-specific risk models to get our views into the portfolio cleanly. It gives us a much more accurate estimate of our tracking error—our distance from the benchmark—and a clear understanding of how our risks are spread across different factors. If we didn’t have this technology, we’d be underestimating risk, running blind to concentrated positions, and failing to efficiently transfer our research insights into the actual portfolio.

Nick Millington, Aberdeen: Simply buying separate Value, Quality, and Momentum ETFs and blending them together rarely yields optimal results. There are a few reasons for this. First, when building a multi-factor proposition using individual single-factor blocks, those portfolios may overlap in unintended or poorly understood ways—unless you have highly sophisticated look-through analytics to monitor the combination. A standalone Value or Quality ETF is optimised for that specific factor; it wasn’t engineered to coexist efficiently within a combined structure.

You get a much more efficient portfolio when you implement a bottom-up integration of all three factors into a single, unified portfolio. Furthermore, many off-the-shelf multi-factor ETFs are not structured relative to standard benchmarks, meaning they often introduce unmanaged tracking error, whereas our approach is strictly risk-controlled.

What triggers drive changes in decision-making within enhanced index investing?

Q: What triggered the shift from overweight to underweight in your semiconductor example? Also, given the macro shifts and rising tensions in the Middle East over the last two to three months, how has the portfolio evolved? Seeing how these decisions play out in detail would be helpful, but the initial concept looks interesting.

Nick Millington, Aberdeen: We don’t just compare all stocks globally on value and quality. If you build a raw value portfolio, you probably end up heavily overweight in sectors like utilities due to their low Price-to-Earnings ratios and high yields. Instead, we compare like-for-like stocks within the same peer group. We are essentially ranking stocks and trying to make money on the spread between similar stocks to each other.

Nick Millington, Aberdeen: The change is entirely systematic. We take financial data and calculate company exposures every single day. However, we implement those changes on a monthly basis. For us, a monthly cycle is the right balance—it gives us the latest insights, prevents the portfolio from drifting into concentrated positions, and enforces a strict discipline of taking profits and cutting losses.

That said, extreme market events like COVID have taught us the value of dynamic risk management. On rare occasions, if we spot a massive, unintended risk exposure intra-month, we will step in right away to neutralise it. I would much rather manage a risk the moment I see it than just hope it works out. But under normal market conditions, it is our regular monthly rebalancing that shifts the stock weights.

Nick Millington, Aberdeen: Regarding the Middle-East conflict, the good thing is we did not need to do a lot of trading when that event happened because we are always looking ahead. Geopolitical risk isn’t new—tensions have been rising globally, we have had the conflict in Ukraine, and we’ve been modelling these types of situations for a while. We learned a lot from how the market revealed its hand during previous short-term conflicts, noting which stocks did well and which did badly. We turned those insights into a sensitivity model for this specific event.

Because we were already monitoring this and keeping the portfolio neutral, we had a very calm period relative to the benchmark when the Middle East conflict started. Of course, as these events continue, you get more data. Our macroeconomics team builds out scenarios around inflation and market impact so we can stay ahead of it.

I know our core factors are working, but we want to know what else is coming down the line. Honestly, this is one of the benefits we provide to our clients. If they come to us worried about a specific risk, we will model it to see if we’re exposed. If we don’t have an alpha view on it, we would much rather just remain neutral.

How do you balance competing investment factors to ensure a robust portfolio framework?

Q: How do you implement momentum within your strategy? Specifically, when momentum indicators reach extreme levels, do you maintain exposure or do you begin to fade the factor?

Nick Millington, Aberdeen: We always want to make sure that the momentum component of our portfolio is contributing but not dominating. We want to give a balance of value, quality and momentum.

One of the worries about momentum is it is great until it is not, which is why when we are building our factors, we try to make them as robust as possible. So when we build our momentum, we also look at industry momentum, reversals, and analyst sentiment.

Nick Millington, Aberdeen: I would say we are moving closer to a risk parity framework, but not in the traditional sense. If you over-engineer risk parity, you just end up running up a lot of turnover for not much gain, and our experience definitely backs that up. Our standard factor weightings tend to give us a really well-diversified portfolio right from the start.

If our monthly rebalancing comes around and we see the portfolio has lost a bit of its alpha exposure—say, we’re running a little low on momentum—we can add a constraint to nudge that exposure up so the portfolio isn’t completely dominated by Value and Quality.

How are investment processes adjusted to accommodate different regional market realities?

Q: You mentioned you have different building blocks geographically. What are the key nuances or differences in how the models operate within each specific market?

Nick Millington, Aberdeen: The process actually does vary a bit by region, particularly within our Quality factor. We differentiate between emerging and developed markets. In developed markets, if a company lacks financial discipline and burns a lot of capital without hitting its return targets, our model flags that as a negative. But in emerging markets, where there is huge growth potential, we actually use a specific ‘quality growth’ factor, which we then balance against strict balance-sheet health and capital structure measures.

There are a few other nuances, too. For financial stocks, we use different metrics simply because bank and insurance balance sheets look different from a standard corporate company. Japan is another interesting exception: in the Japanese market, focusing on free cash flow value works well, but momentum is sluggish. Some people say momentum doesn’t work in Japan at all, but our proprietary version actually does capture alpha there—it is just not quite as powerful as it is in other regions.

How does the prevailing market environment impact the efficacy of this strategy?

Q: Is there a particular type of market environment where you think it’s easy to generate alpha, and conversely, more difficult?

Nick Millington, Aberdeen: The worst type of market environment for us is a ‘growth at any price’ regime, where nothing else seems to work. I would not expect our Value or Quality factors to do particularly well there. Momentum can help if you are able to ride the trend, but when those markets become hyper-concentrated in just a handful of names, outperforming the benchmark becomes a real challenge.

That is exactly where our dynamic risk management process steps in to help us recognise what’s happening. Here is a perfect example: our model generally dislikes companies that burn massive amounts of capital. Right now, there is obviously a staggering amount of capital being spent in the Artificial Intelligence (AI) space. Because of that, we realised we had a slight anti-AI exposure in certain sectors—specifically within software.

We have basically put constraints in place to make sure we aren’t too exposed. So, while we might run an underweight position, we make sure it does not get out of hand. We still believe in our core investment thesis, but we are not going to just blindly double down on everything.

Q: How’s the US portfolio doing this year with this environment?

Nick Millington, Aberdeen: It is tough—there’s no doubt about it. The US is our toughest market, and I will not be the first fund manager to tell you that. We constantly get asked about the US market and mega-cap concentration in general.

I think there is still alpha there. There is still potential to outperform. I just think it’s a bit more challenging in concentrated markets compared to markets that have got more going on.

Is quantitative investing truly a “black box,” or is it more transparent than perceived?

Nick Millington, Aberdeen: We try to be as explicit as we can about what we do, so our investors always know they are getting steady Value, Quality, and Momentum exposure. With the rise of factor-testing tools, more people are starting to manage these exposures closely.

Meanwhile, we also take a glass box investing approach. Look, I am a quant—I love factors, and I love numbers—but I am also aware that there are many different things driving stock prices besides just Value, Quality, and Momentum. Our experience has taught us that you have to understand your portfolio from a few different perspectives.

I can tell you exactly what our factor exposures look like, but I do not have an edge when it comes to predicting macro events. For instance, without deep analysis, how do we know if we are accidentally running an overweight or underweight position on the Middle East conflict, or taking an unmanaged bet on oil exposure, or leaning too heavily into whether AI Capital Expenditure (Capex) is going to boom or bust?

That is exactly why we run an additional dynamic risk management process. We use it to immunise ourselves against those specific risks relative to our benchmark, so that at the end of the day, our returns come exactly from the factors we expect them to.

How is AI used in the investment process?

Nick Millington, Aberdeen: While this specific strategy is not an AI-driven fund, our team is far from new to the space; we have actually been running a proprietary AI alpha model for over eight years, continuously refining it.

The key insight we have uncovered is that AI is not a universal solution; rather, its efficacy depends entirely on how it is deployed. To complement our core process, we have developed a dynamic, conditional alpha model designed to capture uncorrelated excess returns.

The underlying theory is systematic: by mapping current market regimes across multiple characteristics—including factor exposures, volatility regimes, and valuations—the model identifies analogous historical periods. It then constructs an alpha model trained on that mathematically relevant history.

While our primary framework remains anchored in Value, Quality, and Momentum because they persist over long horizons, certain short-term signals work beautifully for a few months or a year before decaying. This broader, regime-dependent AI model allows us to adapt to specific environments and improve our near-term return forecasting.

Furthermore, outside of alpha generation, AI has significantly accelerated our research pipeline. When we back-test new alpha signals or complex portfolio construction methods, AI-assisted coding has reduced our development cycle from two weeks down to just two days. This massive boost to research productivity has exponentially accelerated our learning curve and model iteration.

Where does enhanced indexing best fit within a client’s overall portfolio architecture?

Nick Millington, Aberdeen: For us, we see enhanced Indexing as a core equity portfolio—not too radically different from the benchmark, but with that persistent factor diversification built right into it.

It works well with other overlays. If you have your own tactical regional views or additional factor preferences, our fund serves as a complementary base, rather than trying to force all of those moving parts into this individual strategy.

We inherit the factor exposures of the underlying benchmark because our primary objective is to deliver that standard index return. The outperformance we generate on top comes from our balanced, diversified exposure to Value, Quality, and Momentum.

This strategy is designed to be a beta-neutral, risk-controlled alternative to passive indexing. It does exactly what it says on the tin: it provides highly predictable, persistent risk premium returns. Because our factor exposures are transparent and steady, you can manage your own top-down overlays by mixing our funds with other products or dynamically shifting regional allocations.

Meanwhile, when assessing our value proposition on a fee-adjusted basis, the amount of potential alpha we deliver relative to the cost is substantial.

What differentiates Aberdeen from other asset managers in the enhanced index space?

Q: If competing asset managers are building their own models using the exact same core factors—Value, Quality, and Momentum—how different will their models actually be from yours in practice?

Nick Millington, Aberdeen: Everyone has different recipes for their factors, uses different factor mixes, and decides whether or not to time them over the cycle. For us, the portfolio is king. We focus on a steady-state, robust combination of factors. On top of that, I would say we operate at the lower end of the tracking error spectrum when it comes to our risk budget.

How does the team monitor and control tracking error?

Nick Millington, Aberdeen: We target a persistent and consistent 1% tracking error target over time. While we have the capability to engineer portfolios with higher tracking error parameters, those represent distinct solutions tailored for clients seeking an elevated risk-return profile.

We can outperform at higher tracking error levels. But in my view, there’s a tipping point around the 3% to 4% mark. The issue is that when you increase your risk budget in a long-only portfolio, you inevitably end up taking larger individual stock positions. That means you are taking on a lot more stock-specific risk, which requires a much more concentrated insight into those exact names.

How is quantitative investing utilised by private banks, and what role can enhanced indexing play in DPM?

Paras Gupta, UBP: We manage a specialised quantitative strategy which emphasises the Quality factor and corporate governance as part of its investment process. We prefer to manage the factor-based investment strategies as distinct portfolios, rather than co-mingling them within our core multi-asset portfolios, which follow UBP’s high conviction approach.

Enhanced indexing is a good concept, and it would be interesting to look at three years of baseline data to see how those different factors being taken into consideration interplay with each other.

Robin Chay, DBS: There is a place for this type of strategy. In our portfolios, we always look for active returns and want to show high conviction where we have the expertise. But there are also areas where we may not have that edge, and we want straightforward market exposure. It makes sense to use a cost-effective product with a 1% tracking error there because it’s so broad-based.

Yves Bogni, UBS: We don’t currently utilise these types of enhanced indexing strategies. The way we’re building our portfolios is very targeted. We differentiate between our long-term strategic holdings—where we look for high-conviction active managers—and our tactical allocations. For our tactical trends, we typically use standard ETFs on the equity side.

In our case, this is not an obvious alternative for the ETFs, but it could be a potential alternative for the active managers in certain specific markets, where we see active managers struggling over long periods.

Daniel Furer, Standard Chartered: Given its cost efficiency, enhanced indexing has the potential to replace a portion of traditional passive ETF exposure within portfolios. If implemented successfully, the incremental return generated over time could help improve overall portfolio efficiency while maintaining broad market exposure.