The multiple answers one question exactly - how many times your money came back - and refuses to answer any other. Its blindness to time is a feature and a trap.
10 min read · Level 4
The conventional view is that the multiple is the naive measure and the rate is the sophisticated one, on the grounds that ignoring time is an omission. It is worth considering the opposite reading: the multiple ignores time deliberately, and in doing so it becomes the only return figure that cannot be improved by anything except making more money. Every lever that raises a reported IRR - exiting sooner, distributing earlier, borrowing so that capital is called later - leaves the multiple exactly where it was. That makes the multiple a poor decision tool and an excellent integrity check, and those are different jobs. The practical consequence is a habit rather than a preference: read the rate to decide, and read the multiple to find out whether the rate was earned. Where the two tell consistent stories the figures are probably describing the underlying asset. Where a strong rate sits beside a thin multiple, something in the presentation is doing work that the investments did not.
Multiple on invested capital is total value divided by the money put in. Put in 100,000, get 300,000, and the multiple is 3.00x. There is no time in the formula and none is implied.
In funds the same idea appears in two named forms and the difference between them is the whole of the last lesson in this subject. DPI - distributions to paid-in - counts only cash actually returned. TVPI - total value to paid-in - counts distributions plus whatever the manager currently values the remaining holdings at. A fund at 0.4x DPI and 2.5x TVPI has returned forty percent of the money and is asserting the rest.
And there is gross and there is net. A gross multiple describes what the underlying investments did. A net multiple describes what reached the investor after management fees and the manager's share of the profit. These are routinely far apart, and which one is being quoted is often left to context.
Because a multiple is the one return figure that cannot be improved by timing. Exiting earlier raises an IRR and does nothing to a multiple; deferring a capital call raises an IRR and does nothing to a multiple. Anything that moves the multiple moved the money.
And because private markets have a specific reason to quote it: capital comes back in irregular lumps over a decade, and the single question an investor eventually has is how much money came back for the money they put in. That question has one exact answer and no assumptions in it, which is worth something in a market where almost every other figure has several.
Educational, not personalised financial advice. Nothing in the Investor Academy is a recommendation to buy, sell or hold anything — what is right for you depends on your own circumstances and the rules where you live.
Two deals, each starting with 100,000.
Deal A: returns 300,000 after nine years. Deal B: returns 200,000 after three years.
On the multiple, A wins clearly: 3.00x against 2.00x. On the rate, B wins clearly: A's IRR is 3 to the power of one-ninth, minus one, which is 12.98%; B's is 2 to the power of one-third, minus one, which is 25.99%.
The two measures rank the same pair in opposite orders, and neither is being misused. They are answering different questions: how much money came back, and how hard it worked while it was away.
The tie-break is a question about the world, not about the arithmetic. B gives you 200,000 at the end of year three, and A is still running for six more years. If that 200,000 sits in cash, then at year nine B has produced 200,000 against A's 300,000 - A wins by 100,000 on a common horizon. For B to catch A, the 200,000 has to be redeployed for those six years at a rate of 300,000 over 200,000, to the power of one-sixth, minus one - which is 6.99% a year. Whether that is available, and whether you will actually find something to put it in, is the entire question, and no return measure contains the answer.
Now the gross-to-net gap, which is where multiples are most often misread. An investor commits 100,000 to a ten-year fund. Management fees are 2% of the commitment each year, so 2,000 a year for ten years, which is 20,000 in total, and they are drawn from the commitment. That leaves 80,000 actually invested.
Invested: 100,000 - 20,000 = 80,000 Gross return of 2.50x on that: 80,000 x 2.50 = 200,000 Profit above the 100,000 committed: 200,000 - 100,000 = 100,000 Manager's 20% share of that profit: 20,000 Net to the investor: 200,000 - 20,000 = 180,000 Net multiple on the commitment: 180,000 / 100,000 = 1.80x
A gross 2.50x reached the investor as 1.80x. The gap of 0.70x is 28.0% of the gross multiple, and every input to it was stated in the fund's own documents. This is a simplified structure - real waterfalls add a hurdle rate, catch-up provisions and fee offsets - but the direction and the rough size are representative, and the exercise of doing the arithmetic once is what makes a quoted multiple readable.
The first mistake is comparing multiples without their durations. 3.00x is a superb outcome over four years and an unremarkable one over eighteen, and the number is identical.
The second is comparing a gross multiple with a net one. The difference in the example was 0.70x on a 2.50x, and the two figures look alike on a page.
The third is reading TVPI as a result. Most of a 2.5x TVPI in an immature fund is the manager's own valuation of assets they have not sold, which is a mark and not money.
The fourth is choosing whichever measure flatters. A manager whose multiple is thin will lead with the rate; one whose rate is modest will lead with the multiple. Both figures are cheap to compute and should always be quoted together, which is why a document showing only one is worth a question.
Neither is sufficient and both are trivial once the cash flows exist. A pair - 2.00x over three years, 25.99% - cannot be gamed by presentation in the way either figure alone can.
Net of management fees, net of the manager's profit share, or gross of both. If the document does not say, that is the question to ask before reading any further, because the honest answer can be 0.70x away from the headline.
DPI is what has actually been distributed; the remainder of TVPI is an estimate. Write them as two numbers rather than one, because they are different kinds of claim and only one of them has been tested by a buyer.
Where a multiple and a rate disagree, extend the shorter deal to the longer one's end date and ask what the returned capital realistically earns in between. In the example, B needs 6.99% a year for six years merely to match A.
Whenever you want a figure that timing cannot flatter, when comparing outcomes within a fund or a vintage, and as the necessary companion to any quoted IRR.
Do not use a multiple alone to compare investments of different duration, and do not treat an unrealised multiple as an outcome - a 3.0x mark on something unsold is a valuation, and it is the subject of the last lesson here.
In private-market reporting the most consequential ambiguity is rarely the multiple itself but what it is a multiple of. Invested capital, called capital and committed capital are three different denominators, and a fund that has called 60% of its commitments can honestly report a considerably better figure against invested capital than against what the investor has undertaken to provide. All three are used, none is wrong, and the label 'multiple' is commonly applied to all of them without qualification.
Ask what the denominator is before reading any multiple: invested, called, or committed. The same fund produces materially different figures on each, and the investor's own economics depend on the commitment.
Create a free account to read the rest and apply it to your own investing.