Fee Compression: Why the Margin Falls Faster Than the Fee

Fee compression is a sustained fall in the rate an asset manager charges on the value it holds. The cost line does not fall with it, so the operating margin falls faster than the rate does. At Vaidehi Asset Managers Limited, invented, a blended fee of 0.55 per cent of assets under management would leave nothing at all by the time it reached 0.33 per cent.

The rules on what may be taken out of a pooled arrangement, the base any such charge is struck on, whether a charge tied to an outcome may be taken at all and what a manager has to publish about its charges are settled by the Securities and Exchange Board of India (SEBI) at sebi.gov.in, and each of them moves. A printed figure would not merely go stale; it would be wrong from the morning it changed. The Association of Mutual Funds in India (AMFI) at amfiindia.com is the place to look up an industry total. The Institute of Chartered Accountants of India at icai.org governs when a fee earned across a year is recorded as revenue.

Here is the whole of the arithmetic in one line, before any of the reasoning. A price is being cut on a business whose costs were fixed before the cut. Revenue is a rate multiplied by a value, and only one of those two is falling, so anybody watching the product of them can miss the cut entirely. The profit is what is left after a cost line that did not move at all. So the profit moves more than either of the two numbers that made it.

One distinction carries the rest. A rate is a price and a revenue line is not, and fee compression hides in the distance between those two sentences.

What is actually falling when a fee is compressed?

The rate. Nothing else. Fee compression is a fall in the rate charged on the value held, and the everyday use of the phrase is loose enough that this needs saying plainly before anything is built on it. Compression is not a fall in revenue. Compression is not a fall in operating profit. And compression is not a fall in the rupee amount any one investor hands over in a year. An investor whose holding has grown can hand over more rupees at the lower rate than at the higher one.

Vaidehi Asset Managers Limited, invented for teaching, holds Rs 1,80,000 crore of assets under management and charges a blended feeOne rate that stands in for many. A blended fee is the weighted average of the rates charged across everything a manager runs, so no single arrangement is necessarily charged at it. of 0.55 per cent of assets under management across the year. The blended rate is the price of the service. Everything below is a consequence of it multiplied by something.

A rate that falls while the value it is charged on rises produces a revenue line that does not move at all, and the price of the service has still been cut. The claim sounds like a paradox and is arithmetic. The trap follows from it: the number a reader reaches for first is the one number that has been engineered, by nothing more sinister than multiplication, to say nothing about the thing they wanted to know.

Why does the revenue line say nothing about it?

Because revenue is the rate multiplied by the value, and the value moves for reasons that have nothing whatever to do with the rate. Markets rise. Investors hand more money over. Both of those enlarge the number the rate is charged on, and neither of them is a decision the manager made about its price. So a falling rate can be hidden by a rising value, completely, and for years at a stretch.

Think of a stall that charges two rupees to fill a bottle of water. Trade doubles, and the owner cuts the price to one rupee fifty. The takings at the end of the day are higher than they have ever been, and the price of a bottle of water has fallen by a quarter. Nobody reading the takings alone would ever find that out. The takings are a product; the price is one of the two things multiplied to make it.

To see a fee falling, divide revenue by the value it was charged on and watch that quotientThe answer a division gives. A product is what two numbers make when multiplied together. A quotient is what one number becomes when divided by another. across periods. The quotient is the price and the product is not. The division is a single one, done from two lines of a published result, and it is the only reading here that a person with no access to anything private can perform.

Why does the operating margin fall faster than the fee does?

The faster fall is an identity rather than a tendency, so it is worth proving rather than asserting. Operating profit is revenue less a cost line, and the cost line does not move when the rate moves. Research costs what it costs whether the rate charged is 0.55 per cent of assets under management or 0.33 per cent. So every rupee of revenue given up to a lower rate comes out of the operating profit whole, with nothing between the two to absorb any of it. The property has a name, operating leverageWhat a business has when its spending stays put while its takings move. Profit then swings by more than the takings did, whichever way they went., and it works in the direction nobody enjoys just as reliably as in the direction they do.

The percentage fall in the fee that wipes out the operating profit is exactly the operating margin. Read that twice. The two numbers involved sit on completely different bases and land on the same figure anyway. At Vaidehi Asset Managers Limited the margin is 40.0 per cent, struck on revenue. The fee falling from 0.55 per cent to 0.33 per cent of assets under management is a fall of 0.22 percentage points, and 0.22 over 0.55 is 40.0 per cent of the fee. One is a share of revenue and the other is a share of a rate, and they are the same number to the decimal.

Nothing about that depends on this manager. Write the cost line as any amount at all, write the fee as any rate at all, and the rate has to fall by exactly the share of revenue that the profit represents before the profit reaches nothing. The arithmetic is worth learning once rather than recomputing it nervously each time. The answer is already printed at the bottom of the manager's own results, sitting there as the operating margin.

What is pushing the rate down, and can one manager stop any of it?

Five separate pressures show up again and again, and the useful exercise is not listing them but asking the same question of each one: is this something a single manager decided, and can a single manager undo it?

The first is scale. A larger manager spreads much the same cost line across a larger value, so it can charge less for the same work and still keep something. The second is money moving towards arrangements that cost less to run. The drift drags the weighted average rate down even if no single rate anywhere changes. The third is unbundlingCharging separately for work that used to be sold at one price. Two charges replace one, and the parts become comparable in a way the single price never was.: the work of choosing what to hold is separated from the work of advising a person what to do, so each gets paid for on its own. The fourth is disclosure. Disclosure makes charges comparable where they were once folded together, and a charge that can be compared behaves differently from one that cannot. The fifth is simply more places to buy from, so a person who dislikes a price can act on the dislike.

Not one of those five is inside any single manager's control, and the word structural in the phrase is doing exactly that work. The pressure is structuralDriven by the way a market is arranged rather than by what any one participant chose to do. A single participant can respond to it and cannot switch it off. rather than competitive: it is a property of how the market is arranged and not a move anybody made. A manager can respond to every one of them and cannot switch any of them off, and neither can all of them together.

What can a manager do about a falling rate?

Four responses exist, and reading them as a list of four misses the point. Sort them instead by whose agreement each one needs before it can happen.

Gather more, so the same cost line sits on a larger value: that needs investors to decide to hand money over. Change the mix, offering more of what still carries a higher rate: that needs investors to decide to hold that instead. Change what is charged for, bringing in a charge on the outcome: that needs investors, and often somebody else again, to agree to a different arrangement. And spend less, needing nobody's agreement but the manager's own.

Three of the four responses are decisions somebody else has to make, and one of them is not. Sorting by whose agreement is needed beats sorting by cheap against expensive or fast against slow. The sorting shows which lever gets pulled when the pressure arrives. Under pressure, people reach for what they can move today. So does a business.

Where does a Performance Fee fit when the charge on size is squeezed?

A performance fee is a charge taken on the part of an outcome that sits above a stated reference, rather than on the size of the money held. A performance fee belongs in a treatment of compression for one reason, worth saying out loud rather than assuming: when the rate that can be charged on size is being squeezed, changing what is charged for is one of the four ways out, and a charge on the outcome is the response that changes who is paid regardless.

A charge struck on size is paid whatever happens to the money, and a charge struck on the part of an outcome above a stated reference is not. Only that fourth response changes who is paid regardless. Gathering more leaves the manager paid on size. Changing the mix leaves the manager paid on size. Spending less leaves the manager paid on size. Only the fourth alters the thing the charge is attached to.

The comparison of the two charges criterion by criterion is set out under performance fees. Whether a charge tied to an outcome may be taken for a given arrangement at all, what reference it would be measured against, and what stops the same rise being charged for twice are set by SEBI at sebi.gov.in.

Why is the cost line the only lever the manager controls outright?

Put the four standing questions of this business to a manager watching its rate fall, and the answers arrive in an order that explains the behaviour completely. Whose money is it? Somebody else's. Whose decision is it whether that money stays or leaves? Somebody else's again. Whose risk is it if the value falls? The person whose money it is. And who is paid regardless, at least where the charge is struck on size? The manager.

Now add the fourth question: whose decision is it what gets spent on people, on research and on systems? The manager's, entirely, immediately, and without asking anybody. The asymmetry is the whole reason the cost line is where the pressure lands. Every other response requires an investor to agree with something.

A manager under a falling rate reaches for the one lever it controls outright, and that lever is the people, the research and the systems the service is actually made of, so cutting it is never free. A cost line is not fat; it is the analysts who read the filings, the technology that keeps records straight, and the people who answer when something goes wrong. Cut it far enough and the thing being sold is a different thing. Whether any particular cut goes too far is not a question arithmetic can settle.

What does the arithmetic look like at Vaidehi Asset Managers Limited?

Every figure below belongs to Vaidehi Asset Managers Limited, and every one of them is for a single year. The rate is moved and everything else is held. Holding the rest still is the only way to see the identity rather than take it on trust. Read the table across, and notice that the third column is the one the reader would never have predicted from the first.

Work the crossing point out yourself and the mystery goes out of it. Take the cost line of Rs 594 crore and divide it by the Rs 1,80,000 crore it sits against, and 0.33 per cent of assets under management drops out. Charge exactly that rate and the year brings in Rs 594 crore, the cost line to the rupee, so nothing is left. The crossing point is a break-evenThe point at which revenue exactly meets costs and nothing is left over in either direction. A break-even is a level, not a target and not a forecast. expressed as a rate rather than as an amount, and expressing it that way is what makes it comparable with the fee.

The two numbers that matter most sit on different bases and are the same number: 0.22 over 0.55 is 40.0 per cent of the fee, and Rs 396 crore over Rs 990 crore is 40.0 per cent of revenue. The second is printed in the manager's own results. So the answer to how far the rate can fall before nothing is left was already published, in a different unit, by the manager itself.

What does a falling fee not mean?

Three conclusions are waiting for a reader at this point, and none of the three follows.

A falling fee does not mean the service got worse. A price and a quality are different things, and nothing in the arithmetic above touches what is being delivered for the money. A falling fee does not mean the service got better either. The opposite mistake is just as common, and a lower price is not evidence of a better anything. And it does not mean any particular manager is in trouble. A manager whose value held rose faster than its rate fell has more revenue than it had before, not less, and its margin will have moved by the difference between those two movements rather than by either one.

The arithmetic shows what it shows and no more. The rules on what may be charged and the base any charge is struck on are set by SEBI at sebi.gov.in, and both move. Every rate above belongs to an invented business and was chosen to make a single identity visible.