What is factor investing?

Factor investing is the practice of describing a portfolio by measurable stock characteristics rather than by individual names. A factor is a rule you can compute for every stock, such as how cheap it is or how profitable it is, and a factor portfolio is built by ranking the universe on that rule. The idea is that stocks sharing a characteristic tend to move together and have a common return pattern.

Which factors are most commonly used in India?

The five that appear most often in Indian index and fund documentation are value, momentum, quality, low volatility and size. Indian index providers publish rule based factor indices for each of these, and their methodology documents set out exactly which inputs are used and how the ranking is done.

Does factor investing always work?

No. Every factor has gone through long stretches, sometimes several years, of lagging a plain market cap weighted index. Factor returns are cyclical, and the discomfort of those droughts is part of why the characteristics persist. Anyone using factors should plan for periods of underperformance rather than assume them away.

Factor Investing in India: What Factors Are and How They Are Measured

Factor investing means describing and building portfolios using measurable characteristics that many stocks share, rather than treating every company as a one off story. A factor is simply a rule you can compute for every name in a universe, such as how cheap the stock is, how profitable the business is, or how strongly the price has trended, and a factor portfolio is what you get when you rank the universe on that rule and hold the top slice.

The appeal is that it turns vague investing language into something you can measure, test and audit. “I like quality businesses” becomes a specific set of inputs with a specific ranking method. In India, where index providers now publish rule based factor indices and a growing set of funds track them, factor language has moved from academic papers into ordinary fund documentation.

What a factor actually is

A factor has three parts, and all three have to be pinned down before the word means anything.

A definition. The characteristic has to be computable from data every company in the universe has. “Good management” is not a factor because there is no formula for it. “Return on equity over the last twelve months” is, because you can calculate it for every name and rank them.

A universe. A factor score only means something relative to a defined group. The same stock can rank near the top of a broad market universe and near the bottom of a large cap only universe. Index methodology documents always specify the eligible universe first, and often add liquidity and listing history screens before any ranking happens.

A construction rule. Once every stock is scored, you need to say what happens next. Do you take the top fifty names? The top quintile? Do you weight them equally, by market cap, or by the factor score itself? How often do you rebalance? Two portfolios claiming to follow the same factor can behave quite differently because these choices differ.

That third part is where most of the confusion lives. When two “value” funds behave differently, it is usually not because one has a secret insight. It is because one ranks on price to book, the other on a blend of earnings and cash flow yields, and they rebalance on different calendars.

A factor is a definition plus a universe plus a construction rule. Change any of the three and you have a different portfolio, even if the label on the front is identical.

The factors that show up most often

Five characteristics dominate the published Indian factor menu, and each has its own article in this series.

Value ranks stocks by how cheap they look relative to something fundamental: earnings, book value, sales or cash flow.
Momentum ranks stocks by their own past price trend over a lookback window, usually with the most recent month skipped.
Quality ranks stocks on profitability, earnings stability and balance sheet strength.
Low volatility ranks stocks by how much their prices bounce around, favouring the calmer names.
Size sorts by market capitalisation, with the smaller end treated as the factor tilt.

Others appear in institutional work: dividend yield, growth, and various liquidity or leverage tilts. But these five are the ones you will meet in Indian index methodology documents and in most fund literature.

Why anyone believes factors mean something

There are two families of explanation, and honest practitioners hold both loosely.

The first is compensated risk. On this view a factor earns something over time because holding it is genuinely uncomfortable. Cheap stocks are often cheap because the business is struggling or the sector is out of favour, and you are being paid for bearing the chance that the trouble is real. Smaller companies are less liquid and more fragile, and the return is compensation for that fragility. Under this reading, the factor is not a free lunch. It is a risk you have chosen deliberately.

The second is behavioural. On this view certain patterns persist because investors systematically over react or under react. Momentum is usually explained this way: news gets absorbed into prices gradually, and investors chase what has already worked. Low volatility is often explained by the preference many investors have for exciting, high beta stories, which leaves the boring names relatively unloved.

Both explanations have supporters, and the practical implication is the same either way. If the factor exists because of risk, you must be willing to bear that risk through bad stretches. If it exists because of behaviour, it only persists as long as enough investors keep behaving that way, and you must accept that a well known pattern can weaken as more money chases it. Neither story guarantees a future return.

Measuring factors in the Indian context

Two practical realities shape factor work on Indian equities.

Data depth is uneven across the market. Large companies have long, clean, continuously reported histories. Move down the market cap ladder and reporting is thinner, restatements are more common, and history is shorter. A factor score computed on ten years of data for a large cap and on three years for a small cap is not the same measurement, even if the formula is identical.

The data must reflect what was knowable at the time. A factor study is only meaningful if each stock’s score on a past date used the numbers that were actually published by then. Annual results arrive months after the year ends, and companies restate prior periods when accounting standards change or a business is demerged. Scoring history with today’s restated figures quietly hands the model information from the future. This is the core of point-in-time discipline and of lookahead bias, and it is the single most common reason a factor study looks better on a spreadsheet than in a live portfolio.

Corporate actions matter too. Splits, bonuses and rights issues change the raw price series, and momentum or volatility computed on unadjusted prices will produce nonsense at exactly the wrong moments.

The risks

Factor investing has real and well documented downsides, and skipping over them is how people end up abandoning a strategy at the worst possible time.

Long droughts are normal, not exceptional. Every major factor has gone through multi year stretches of lagging a plain market cap weighted index. Value in particular has had famously long periods out of favour in markets around the world. A three year drawdown against a benchmark is not evidence the factor is broken, and it is also not evidence it will recover. It is simply what factor cyclicality looks like from inside, and it tests conviction in a way a spreadsheet never does.

Definitions are choices, and choices can be tuned. Because a factor is a formula, it is easy to keep adjusting the formula until the historical numbers look attractive. That is overfitting, and it is the most common way factor research fools its author. A definition that needs three specific parameter values to work is fragile.

Implementation eats returns. Factor portfolios rebalance, and rebalancing costs money in brokerage, taxes, slippage and market impact. Higher turnover factors such as momentum are hit hardest. A paper result computed on closing prices with no costs is not a result.

Crowding changes behaviour. When a lot of capital tracks the same rules, the stocks entering and leaving those portfolios can move on rebalance dates for reasons unrelated to their businesses. That is factor crowding, and it can compress the very edge people are chasing.

A factor label is not a business assessment. A high quality score says the ratios currently rank well. It says nothing about whether a competitive position is durable, whether accounting is conservative, or whether management is honest. Factor scores are a filter for attention, not a substitute for reading the filings.

What it does not tell you

A factor score is a ranking within a universe on one measurable characteristic at one point in time. That is all it is.

It does not tell you why a stock ranks where it does. A stock can enter a value portfolio because the market has overreacted to temporary bad news, or because earnings are permanently impaired. The score cannot separate the two, and the difference is everything.

It does not tell you what else you own. A portfolio built on one factor almost always carries unintended exposure to others, and to particular sectors. A quality tilt often ends up concentrated in a handful of sectors. That is why factor exposure analysis exists as a separate discipline.

It does not tell you the right time to use a factor. Factor timing is genuinely difficult, and the honest position among practitioners is that most attempts at it fail.

And it does not tell you anything about a specific stock’s prospects. A factor is a statement about a group, computed across many names over long horizons. Nothing about the group average carries over to a claim about any individual holding.

Portfolio metrics explained: the hub for the risk, return and factor measures referenced here.
Multi factor investing explained: how several factors are combined into one portfolio, and the trade offs in doing so.
Factor cyclicality and drawdowns: why droughts happen and how teams size for them in advance.
What is ROCE: one of the profitability inputs that commonly feeds a quality score.
Why point-in-time data matters: the data discipline any credible factor study depends on.

This article is educational. Altys Labs is not a registered research analyst or investment adviser, and nothing here is investment advice or a recommendation to buy, sell, or hold any security.