Measuring Portfolio Drift: How a Portfolio Wanders From Its Mandate

Portfolio drift is the slow, usually unintended movement of a portfolio away from what it was meant to be: its stated style, its size profile, its sector balance, and its concentration limits. You measure it by fixing a baseline, either the fund’s mandate or the way the portfolio was originally designed, and then comparing today’s actual exposures against that baseline across a handful of dimensions. The distance between the two is the drift, and the whole point of measuring it is to catch that distance while it is still small.

Drift is dangerous precisely because nobody decides to do it. A manager does not wake up one morning and choose to become a technology fund or a mid-cap fund. It happens one price move and one small position at a time, and by the time it is obvious in the returns, the portfolio an investor actually holds is no longer the portfolio they were sold.

Why a portfolio moves even when you do nothing

The first thing to understand is that drift needs no action from you at all. Weights change on their own.

Suppose you build a portfolio with equal weights across ten holdings. A year later, three of them have doubled and two have halved. You have not traded a single share, but your portfolio is now a completely different shape. The winners are a much larger slice than you intended, and the losers have shrunk into the corner. If those winners happen to sit in the same sector, or share the same style, the whole portfolio has quietly tilted toward that sector or style without a single deliberate decision.

This is the counterintuitive part. Doing nothing is itself a choice, and in a portfolio it is a choice to let the market set your weights. Price movement is the largest source of drift, and it is the one people forget, because it does not show up as a transaction anywhere. On top of it sit the smaller, more visible sources: new purchases that lean the same way, forced selling to meet redemptions, and a long series of individually reasonable position decisions that add up to a direction nobody chose on purpose.

The four kinds of drift worth measuring

Drift is not one number. It is several different movements, and they matter for different reasons. Four are worth tracking on their own.

Style drift is a move along the value-to-growth axis. A fund that describes itself as value-oriented, meaning it leans toward companies trading at lower multiples of earnings or assets, can slowly fill up with faster-growing, higher-multiple names, either because those names ran up or because they were bought. The label still says value. The holdings increasingly do not.

Size drift is a move along the market-capitalisation axis, from large-cap toward mid or small-cap, or the reverse. This one matters a great deal in an Indian context, where fund categories are defined by size bands. A portfolio that was comfortably large-cap can drift toward mid-cap simply because its mid-cap holdings outran the rest, and mid and small-caps carry very different liquidity and volatility than the large-caps an investor may have expected.

Sector drift is a change in how weight is spread across industries. A fund that was balanced across financials, consumer, industrials, and technology can, through nothing but relative performance, end up heavily concentrated in whichever sector led the market. The portfolio now carries a sector bet that was never explicitly placed.

Concentration drift is a change in how much sits in the top few names. As winners compound, the top ten holdings can swell from a comfortable share into a dominant one, so the portfolio’s fate rests on a shrinking number of positions. The reverse also happens, where a portfolio slowly spreads so thin that no single name can move it.

How to actually measure it

Measuring drift is a comparison, and a comparison needs two things: a baseline and a current snapshot. Get the baseline right and the rest is arithmetic.

1. Fix the baseline. This is the reference you are measuring against. It can be the stated mandate (the category rules, the style the fund advertises, any concentration or sector limits it commits to), or it can be the portfolio as it looked on a chosen start date, or a benchmark index that represents the intended universe. The baseline is a deliberate choice, so write it down. Drift is meaningless without stating what you drifted from.

2. Snapshot today’s exposures on the same dimensions. For the current portfolio, compute the same set of measures the baseline is expressed in: the weighted split between value and growth characteristics, the weighted split across market-cap bands, the weight in each sector, and the share held in the top five or ten names.

3. Compute the gap on each dimension. Line the two up and take the difference. A simple, honest way to summarise sector or size drift is the sum of the absolute differences in each bucket’s weight, which tells you, in one number, how much of the portfolio would have to move to get back to baseline.

Dimension	Baseline	Today	Gap
Value vs growth tilt	Value-leaning	Growth-leaning	Style has crossed the line
Large / mid / small split	80 / 15 / 5	60 / 30 / 10	20 points migrated down in size
Top-10 concentration	35%	52%	17 points more concentrated
Largest sector weight	22%	38%	16-point sector build-up

The numbers above are illustrative, not from any real fund, but they show the shape of the answer you are looking for: not a vague sense that the portfolio “feels different”, but a specific, per-dimension reading of how far it has moved and in which direction.

Drift is undramatic by nature. No single quarter looks like a decision. It is only when you compare against a fixed baseline that the accumulated move becomes visible at all.

Turnover and active share: the companion readings

Two related numbers help you interpret drift, and it is worth being clear on what each one does, because they are easy to confuse.

Turnover measures how much you traded over a period. It answers “how busy was the portfolio”, not “how far did it move”. This distinction matters because a portfolio can drift a very long way with almost no turnover, purely because prices moved the weights, and it can also churn heavily while staying exactly on-mandate. If you see large drift alongside low turnover, the market did it to you. If you see large drift alongside high turnover, the decisions did it. Same drift, very different story.

Active share measures how different a portfolio is from its benchmark, expressed as the share of holdings that do not match the index. On its own it is a snapshot of distinctiveness. Watched over time, a change in active share is itself a form of drift: a portfolio slowly converging toward its benchmark is becoming less active, which is worth knowing whether or not it was intended.

Read together, these turn drift from a single static gap into a story with a cause. That is the same instinct behind the thesis monitoring checklist: a reading is only useful once you know what moved it.

Why the measurement matters

The reason to do any of this is that undeclared drift is a mismatch between the risk an investor believes they hold and the risk they actually hold. A person who chose a large-cap, value-leaning fund for its steadiness has a right to expect it to stay roughly that. If it has quietly become a concentrated, mid-cap, growth-tilted portfolio, its behaviour in a downturn will not be what they signed up for, and they will discover the change at the worst possible moment.

None of this is about whether the drift was good or bad in return terms. Some drift is the honest result of deliberate active management, and letting strong holdings compound is a legitimate approach. The problem is never movement itself. It is movement nobody measured, described, or decided on. Measuring drift is what lets you tell a considered active bet apart from an accident of price. It sits naturally alongside monitoring a portfolio of holdings and, at the level of a whole book, comparing fund-manager portfolios at scale.

What to take away

Drift is a discipline of comparison, not a feeling. The habit is simple to state and easy to skip.

• Write down the baseline first: mandate, start-date portfolio, or benchmark. You cannot measure drift from a reference you never fixed.
• Measure on four axes: style, size, sector, and concentration. Each drifts for its own reasons and each carries its own risk.
• Read turnover next to drift, so you know whether the market moved you or you moved yourself.
• Treat drift as information, not a verdict. The goal is to catch undeclared movement early, not to prevent every deliberate tilt.

A portfolio is a living thing, and a living thing moves. The job is not to freeze it. The job is to always know how far it has wandered from what it was meant to be, so that any drift left in place is drift you chose. This is part of the wider habit of continuous research and of how PMS firms research Indian stocks at the portfolio level rather than one name at a time.

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.