Factor Models at 50: Innovation that Changed Investing
“We stand now at the edge of the next wave of innovation, creating new applications powered by generative AI, intended to bring even greater clarity to systematic investing.”
Each year major league scouts descend on Cape Cod in Massachusetts to hunt future talent among top collegiate baseball players gathered from across the country to compete in a premier summer baseball league. It is a decades-long tradition that has produced many major league stars.1
Scouts don’t just rely on intuition — they use statistics and performance metrics to evaluate players, scouring batting averages, on-base percentages, pitching velocities and strikeout ratios to identify the best candidates for the professional stage.
So what does this have to do with investing?
Just as scouts identify raw talent who may become major league standouts, MSCI transforms abstract economic and financial research into practical tools for investors. By rigorously testing academic theory against real-world market data, we identify the ideas which may best support our clients in developing investable strategies. And our role in advancing the factor model is perhaps the clearest example of this process.
With the launch of the industry’s first factor model 50 years ago through Barra,2 MSCI has been central to its evolution ever since. What began as a narrow equity framework has since expanded into more than 70 equity factor models covering over 90,000 securities, 49 industries and 85 countries.3
By today’s standards, the first factor model was a basic factor structure and relatively simple risk methodology. But even then, it offered investors a new way to understand the forces driving risk and return — a goal that continues to shape our innovation today.
We stand now at the edge of the next wave of innovation, creating new applications powered by generative AI, intended to bring even greater clarity to systematic investing. And it is the decades of exploration that make this future possible.
Factor investing aims to capture exposures to systematic risk premia — characteristics such as value, low size, low volatility, high dividend yield, quality and momentum — that historical research has shown explain variation in returns and offer long-term premium.4
This evolution hasn’t occurred in a vacuum. Investors have faced increasingly complex challenges with shifting macro regimes, periods of heightened volatility and fluctuating policy-driven risk.
Throughout these shifts, models have helped investors bring greater structure and insight to their decision-making process. The purpose has always been clear: to help clients understand what’s driving their portfolios — and what might lie ahead.
Through the years, Barra models gained traction among institutional investors by offering a standardized framework for risk attribution and performance analysis. This helped clients gain a clearer understanding of the drivers of portfolio outcomes and laid the foundation for more consistent, data-informed investment decisions.
The integration of Barra into MSCI in 2004 marked a significant milestone, embedding factor analytics into MSCI’s broader index and data ecosystem. This development enabled us to deliver more comprehensive solutions, combining index construction with robust risk modeling.
The manifestation of this integration is best seen in our investable factor indexes, used commonly among asset managers for product creations. MSCI’s multi-factor indexes introduced optimization techniques that balanced turnover, exposure and, later, sustainability goals, making factor investing more accessible and practical for a wider audience.
“Today, clients combine our models and indexes, sustainability and climate datasets, stress testing frameworks, economic exposure tools and scenario analysis to more holistically address portfolio challenges. ”
Today, clients combine our models and indexes, sustainability and climate datasets, stress testing frameworks, economic exposure tools and scenario analysis to more holistically address portfolio challenges. These capabilities are increasingly being used in tandem to support total portfolio design, from strategic asset allocation to ongoing monitoring.
MSCI has continued to innovate, introducing a broad dataset called FactorLab, integrating macro risks, alternative data, machine learning and crowding considerations which were later added into models — with the goal of helping clients better detect early market signals, avoid unintended exposures and refine their investment processes.
Equally, the use of this data and models among clients has also evolved. Originally, risk models were used among risk or middle-office teams primarily for governance and risk reporting, but less among portfolio managers and investment teams.
Today, quantitative investment managers use models for security selection, portfolio construction, risk hedging and more. For example, it’s common among investment team clients to decompose a portfolio across factor and idiosyncratic exposure and hedge out unwanted exposures, which could include the style factors, while leaving the stock specific elements to drive investment risk and return.
Other managers use models to ensure the exposure to targeted factors is consistent with their objectives, while shorting or hedging unwanted factor exposures.
Equally important is the systematic risk management trend or “risk to the front office” transition. In this context, risk managers influence capital allocation and rebalancing decisions by collaborating with portfolio managers to more precisely allocate capital toward their strategy objectives. This trend has gained favor among multi-manager hedge funds and, increasingly, within the long-only community with diverse investment teams responsible for parts of a portfolio allocation.
However, factors are not only a tool for quantitative managers. They also provide a framework for managing risk, expressing investment views and navigating market regimes — making them relevant to both fundamental investors and quantamental investors, who combine systematic quantitative models with traditional fundamental research. The “risk to the front office” trend is increasingly evident across both quantitative and fundamental investment firms.
“With the expanding use cases, risk has begun to serve a function beyond compliance, becoming a source of critical insight.”
With the expanding use cases, risk has begun to serve a function beyond compliance, becoming a source of critical insight. For clients, this means the ability to align portfolios with strategic objectives, respond more dynamically to market events, and better identify exposures that may otherwise go unnoticed. At a segment level:
Asset owners can use factor models for due diligence, investment reporting and to align allocations with liabilities, helping to provide transparency into what really drives long-term portfolio risk.
Wealth managers use them to evaluate ETFs and other funds by their true exposures, not just their labels. They design portfolios with deliberate tilts and monitor them for drift over time, often expressing exposures directly through MSCI factor indexes.
Asset managers use factors to build strategies with more precision. They create targeted or multi-factor products, hedge out unintended exposures, and integrate sustainability or climate constraints while preserving factor intent.
Hedge funds take factor models in yet another direction — using them daily to decompose returns, isolate alpha, hedge macro risks and avoid crowded trades.
Much like the first wave of baseball statistics gave scouts a structured way to compare players, early factor models broke risk into clear, measurable components. And just as baseball analytics became more sophisticated, our models over the years have matured, expanded and become widely-adopted across the investment ecosystem.
So where do we go from here? The next chapter will be heavily influenced and perhaps written by generative AI.
Investors regularly face market events that threaten to shift the risk of their portfolios. But they often lack a deeper level of visibility into the drivers behind the risk, which may not be fully captured by existing models.
Clients who recognized how profoundly COVID impacted portfolio risk and return, for example, commonly requested a dedicated “COVID factor.” In traditional factor models, much of the pandemic’s impact appeared as “stock specific” risk, or unexplained variation in returns not attributed to broad style or industry factors.
Divergent sector shifts among airline, technology and healthcare companies were not fully captured by existing factors such as value, size or momentum. Instead, they showed up as idiosyncratic movements unique to each company. And statistical models often flagged heightened volatility without explaining the underlying drivers.
This limitation left investors without the necessary transparency to understand how a single global event was influencing portfolios across sectors. Generative AI may take the next step: measuring the financial relevance of unstructured data and connecting it directly to security and portfolio risk.
While our machine learning factor introduced non-linearities into our linear factor models, AI may help capture the financial relevance of market news and events on a security, and by extension, a portfolio. That is an exciting frontier, which may allow investors greater transparency into the drivers of investment risk and return and further improve capital allocation decisions.
Early factor models demystified portfolio risk by breaking it down into components. AI can potentially help demystify quantitative investing itself, making it less intimidating and more transparent. By explaining the drivers of familiar factor returns and translating complex signals into plain language such as earnings quality, policy sensitivity and valuation shifts, AI has the potential to broaden adoption and help resolve the “black box” criticism that has sometimes clouded quantitative investing.
Our mission is clear: We aim to give investors the tools they need to see what truly drives risk and return so they can quickly adapt to market shifts, move with conviction and build more resilient portfolios for the future.
Subscribe todayto have insights delivered to your inbox.
Portfolio Management
Discover how MSCI equips portfolio managers with tools to navigate market disruption and build more resilient, adaptive portfolios.
Equity Factor Models
See how we help clients build, implement and measure factor-based strategies to better understand the drivers of risk and return.
Factor Indexing Through the Decades
Factor strategies have persisted through market cycles, but macro shocks and new business models are reshaping how they’re built and understood.
1 Source: Cape Cod Baseball League. As of September 2025.
2 Barra, founded in 1975, was acquired by MSCI in 2004.
3 As of March 2025.
4 https://www.msci.com/research-and-insights/blog-post/what-is-factor-investing
The content of this page is for informational purposes only and is intended for institutional professionals with the analytical resources and tools necessary to interpret any performance information. Nothing herein is intended to recommend any product, tool or service. For all references to laws, rules or regulations, please note that the information is provided “as is” and does not constitute legal advice or any binding interpretation. Any approach to comply with regulatory or policy initiatives should be discussed with your own legal counsel and/or the relevant competent authority, as needed.