Franklin Templeton’s $98-billion quant brain lands in India

Adam Petryk, Executive Vice President and Head of Franklin Templeton Investment Solutions

Franklin Templeton is doubling down on data-driven investing in India with the launch of its new Franklin India Multi-Factor Fund (FIMF)—a quant-led equity scheme designed to blend systematic models with human judgment. Backed by the firm’s global quantitative engine that manages over $98 billion, the fund evaluates India’s top 500 companies through a proprietary QVSA framework—Quality, Value, Sentiment and Alternatives—drawing on more than 40 metrics to build a balanced, diversified portfolio.

Moneycontrol spoke with Adam Petryk, Executive Vice President and Head of Franklin Templeton Investment Solutions, who has been with the firm for nearly three decades and has worked on quantitative strategies since 1997. In a deep-dive conversation, Petryk explains why multi-factor strategies work better, how the fund adapts across market cycles, and how the quant model is built — and why it works.

1) What does this fund really achieve?
We take a diversified view of how to identify the companies we want to invest in, how to put them together into the portfolio, and ultimately how to achieve the goal of delivering outperformance for investors. We are launching these India-based strategies, but they are part of a global platform of capabilities that we run around the world. We’ve been doing this globally for our clients, and it’s been a really important driver of our success.

2) Is this ‘systematic active’ approach really working worldwide?
It is. One thing we’ve learned is that ‘factors’ or models can be applied around the world, but you need to tailor the model for the different markets and regions you’re operating in. Our approach does that by design. We have a global capability, but then we look at things within a sector or within a country and tailor the model. That’s how you deliver consistency of results.

3) What is the part that you tailor or customise for a particular region?
The tailoring helps identify the actual factors that are working in a particular market and how you compare the companies you’re investing in within that market.

4) So, the natural question is: what factors work in India—or are working here currently?
Different factors work at different points in time. If your approach relies on only one factor, you’ll have periods of fantastic performance followed by significant drawdowns. We see that in India as well. You benefit from a diversified approach across factors.

In India specifically, we’ve seen that a combination of sentiment factors (more momentum-type factors—companies that have been doing well continue to do well) together with companies that have strong fundamentals, attractive valuations, and attractive earnings profiles works well. So instead of relying on a single factor, combining multiple fundamental styles and behavioral styles gives a very powerful combination, and has worked well in India.

5) How is this fund really different from a simple flexi-cap fund with a broad, diversified mandate?
Even within a diversified fund, if you have a single-factor or style-based approach, you’ll see periods of fantastic outperformance followed by cycles of significant underperformance. To achieve long-term success, you need to weather those ups and downs. By combining multiple styles in one fund, you end up with much more consistent performance. That helps on two fronts: a better information ratio (more return for the risk taken) and investors are more likely to stick with a strategy that delivers consistent returns.

6) So you’re saying it doesn’t sharply outperform or underperform but delivers consistency of moderate performance…
It’s very challenging—even for professional portfolio managers—to time when growth or value will work. If you look over the years, the ranking of factors that perform best changes frequently. Instead of trying to predict which one will do well this year, a diversified approach gives more consistent results.

7) At every point in time, is there a fixed proportion allocated to different factors, or is that driven by market conditions?
The primary philosophical approach is diversification. We tailor the model based on the investable universe. We have a global model that scores 8,000 companies with certain allocation weights. Within each market, we pivot structurally based on how the different factors perform in that market. We do not time the market each year. There is always exposure to all four pillars: value, sentiment, quality, and alternative.

8) Are these the only four factors?
These are the four main pillars. Within them we have more than 40 sub-factors. Quality has more than 10, value around 13–14, sentiment more than eight. The sub-factors help us define what quality (or any pillar) actually means, because these concepts are subjective. We put a number to them in a broad, structured, and experienced way.

This fund is a byproduct of a capability we’ve been running globally for close to 30 years. The model has evolved over time; new factors get introduced when they add information. One area we’ve focused on recently is innovation and AI.

9) How do you use AI in this model?
There are two ways to think about it: first, techniques we can use inside the model (we’ve been using AI techniques for a long time; quantitative investing is AI investing—it’s just more mainstream now); second, how we identify companies that will benefit from the AI revolution. We have factors that look for innovation in company fundamentals. That’s something we’re quite excited about and that differentiates us from competitors.

10) Tell us about your investing experience and beliefs.
I started my career doing both fundamental and quantitative investing. I’ve always believed that’s a powerful combination: investors with a fundamental mindset but robust quantitative techniques. Some models are black-box—you don’t understand what’s happening. Our philosophy is to understand the factors and keep them fundamental. If you’re a fundamental investor, you look at valuation, cash flows etc—these are fundamental factors. We take those, build the factor, and the quantitative model evaluates them every day across a broad universe. In many respects, we’re doing what a fundamental team does—just more often and more consistently.

11) So, the first filter is done through fundamental analysts?
The first layer is the research layer where we identify and evaluate a new factor: Is the factor understandable? Does it make sense? Does it add something new to the model?

12) For each factor, do you backtest and see the effectiveness of its contribution?
That’s correct. We also continue to assess the factor in real time to ensure it behaves as expected.

13) Once you identify the factor and put it in the model, the entire universe is ranked. What about selection and weighting?
Weighting of factors is based on quantitative statistics plus qualitative assessment: the factor’s risk and return characteristics, its correlation with other factors, and human judgment. That’s where an experienced team adds value.

14) Can you tell me some sub-factors that uniquely work for India—or work particularly well here—but may not work elsewhere?
One example is earnings momentum, not just price momentum. We track what brokers are saying today versus three months ago—any upgrade or downgrade. Price momentum can take three to six months to reflect a regime shift, but earnings revisions can capture it earlier.

15) Isn’t it the other way around? Don’t prices usually move ahead of earnings?
Not necessarily. Investors tend to under-appreciate new information. It doesn’t matter what the absolute level of earnings or analyst ratings is—it’s whether they are improving or deteriorating. A company with low ratings that are improving can be a better investment than one with high ratings that are starting to decline.

Sometimes the fundamentals start turning before prices fully reflect it, especially when valuation is attractive. Price momentum and valuation are complementary: momentum keeps you invested in winners even when they become expensive; deteriorating fundamentals trigger an exit. The discipline of using both—and removing human emotion—is very powerful.

16) In a sense, the trigger is pulled by the model?
The primary signal is driven by the model. The portfolio manager still looks at the portfolio as a whole, but the model is the dominant driver. We bring technology and discipline, and the portfolio manager ensures data hygiene, captures lags, news sentiment, or governance issues when needed.

17) Any other sub-factor that is a key differentiator?
Innovation. We quantify how much companies are investing in AI and new technologies across sectors—healthcare, autos, everything. R&D is treated as an expense in accounting and hurts current earnings, but we view it as investment in future growth. We can measure quantitatively whether it is effective, and it has been a significant performance driver.

18) Any sub-factors that are counterintuitive or uniquely Indian?
We’re careful about factors that are truly unique. In the US, a more mature, tech-heavy market, quality works better. India is very bank-heavy, so value and sentiment work better. That’s why weights are tailored to the country and its maturity.

The beauty is that the combination of behavioural and fundamental factors is durable across markets. India is a growth-centred economy, and in such markets these models work particularly well because momentum and improving fundamentals tend to persist alongside economic growth.

19) In this strategy, how much is quant-driven and how much is human intervention? Can you put a percentage on it?
If you run portfolio attribution, well north of 90% is driven by the model. The rest is ensuring the model is doing what it’s designed to do and that the portfolio captures it properly.

This model is like an internal ChatGPT we’re now bringing to India. Sometimes, I step in on data lags, news sentiment, or governance issues. Enhancing the model by one point impacts all 500 stocks, not just one or two. Most Indian quant funds are run by fundamental teams where quant is secondary. Our global team has been doing only quant for 30+ years. We’re crossing $100 billion in quant AUM globally this month, and Indian investors now get access to that IP.

20) What is the data set you use? How many companies under coverage?
In India, we cover 550–600 companies, but the fund uses the Nifty 500 universe for cleanliness and liquidity. Globally, we cover over 8,000 stocks—all in the MSCI All Country World Index.

21) Globally, what proportion of quant-based funds outperform compared with human-managed funds?
At the highest level, both can do well in absolute performance. But quant strategies almost always deliver better risk-adjusted returns because of higher consistency. Quant portfolios are more diversified (often 100+ names versus 30 for fundamental) and use sophisticated risk models and optimizers during construction. The risks we take are intentional; we explicitly balance multiple factors. Fundamental managers typically don’t think that way. For investors who want consistency of outperformance, quantitative strategies have a strong advantage.