John Lukovich
Portfolio Manager, Factor Investing
IMCO
Institutional Connect is pleased to feature a speaker spotlight with John Lukovich, Portfolio Manager, Factor Investing at IMCO. In this interview, John shares insights from his experience managing global multi-factor equity portfolios at IMCO, discussing how factor models have evolved over the past decade and the innovations now essential to systematic investing. He reflects on his transition from asset management to the asset owner side, the trade-offs involved in balancing alpha, diversification, capacity, and transaction costs, and offers perspective on how technology, data, and AI are reshaping the investment landscape, along with advice for young professionals looking to stay relevant in the years ahead.
Institutional CONNECT: You have been involved in the global multi-factor equity portfolios at IMCO. How have factor models evolved in the last decade, and what innovations or techniques are now essential for staying competitive in systematic investing?
John Lukovich: Over the past decade, factor investing has shifted from relying on a handful of well known styles to building diversified, next generation multi signal frameworks. What’s changed is less the headline factors and more how we measure, combine, and risk manage them. Three developments stand out:
- Richer data & localized signals: Alternative data and improved disclosures have made it possible to measure fundamentals, sentiment, and risk more frequently and with better regional nuance—especially valuable in emerging markets where reporting standards vary.
- Modern toolkits: NLP, machine learning, and AI help extract information from text, unstructured sources, and complex interactions. Used thoughtfully, they improve feature engineering, robustness testing, and signal decay monitoring—without turning the portfolio into a black box.
- Portfolio construction & implementation: The edge increasingly comes from how you translate signals into trades: purification of factor exposures, tighter risk guardrails, realistic capacity controls, and predictive transaction cost modeling. That’s where alpha survives after costs.
The net effect is greater dispersion across managers. Those slow to adopt new methods are falling behind, while proprietary research and localized insights are making the space ripe for innovation. For investors, that raises the stakes on manager selection and true diversification across complementary approaches.
John Lukovich: I wanted to work where portfolio construction, implementation, and governance meet. On the asset owner side, you design strategies that fit a long horizon, total portfolio context—balancing return, risk, liquidity, and cost for beneficiaries.
The surprise was how much of the edge comes from plumbing: data quality, model robustness, capacity discipline, and the day to day mechanics of implementation with external partners. You spend more time ensuring the process is resilient across regimes, not just finding the next signal. The upside is the ability to build durable, “all weather” portfolios and to steward innovations—new factors, new data, new techniques—when they are truly additive and scalable.
John Lukovich: I think in layers:
- Start with durable edges: Emphasize well understood factors (value, momentum, quality, etc.) that have worked across time and markets, then diversify their definitions (multiple signals per theme) to reduce model risk.
- Design for all weather resilience: Accept that factor payoffs are cyclical; reduce path dependence via balanced exposures and guardrails that minimize unintended country/sector/beta tilts.
- Understand capacity: Strict AUM and liquidity controls, security level weight and risk caps.
- Trade what you can earn: Predictive TCA, turnover budgets by signal half life, and adaptive rebalancing (e.g., urgency when short lived alpha is strong; patience when costs are high) preserve net returns.
- Allocate risk, not just capital. Track marginal contribution to active risk and cost at the signal and position level; re weight to the highest information ratio opportunities after costs.
In emerging markets, these trade offs are magnified: higher dispersion and structural inefficiencies can raise alpha potential, but country risk, geopolitics, and varying ESG regimes demand tighter controls and localized research.
John Lukovich: Be T-shaped: develop deep domain expertise (accounting, market microstructure, risk, portfolio construction) and broad technical fluency (data engineering, NLP/ML/AI, experimentation). A few principles:
- Use AI as leverage, not autopilot: LLMs can surface more domain knowledge than a typical expert—but only if you ask good questions and supply context. Don’t outsource judgment; use AI to stress test, prototype, and accelerate.
- Obsess over data quality: The biggest alpha leak is bad data or weak labeling. Build habits around lineage, versioning, diagnostics, and reproducibility.
- Think in costs and constraints: Always ask — Can this scale? What will it cost to trade? Implementation discipline is part of the edge.
- Stay curious and humble: Markets adapt. Continuously monitor signal decay, revisit assumptions, and be willing to retire features that no longer earn their keep.
- Communicate clearly: The best ideas win only when they’re explainable to non quants—clarity builds trust and speeds adoption.
