Time-Weighted vs Dollar-Weighted Return

Two return measures answer different questions

Time-weighted return (TWR) measures investment performance while reducing the effect of external deposits and withdrawals. Money-weighted return (MWR), commonly calculated as an internal rate of return (IRR), reflects the size and timing of those cash flows. TWR is useful for evaluating a manager’s investment decisions against a benchmark. MWR better describes the investor’s experience with money invested at particular times. The two can differ even when they use the same portfolio values.

The distinction matters because a large contribution before a strong return changes the investor’s dollars earned, while a portfolio manager may not control when the client contributes. TWR links subperiod returns geometrically. MWR finds the discount rate that makes the present value of contributions and withdrawals, including ending value, balance. Neither is inherently the “true” return for every question; choose the measure that matches the decision.

How time-weighted return works

To calculate TWR, divide the measurement period at each external cash flow. Calculate the portfolio’s return for each subperiod using values immediately before and after the flow, then compound the subperiod returns. If the first interval returns 10% and the second returns −5%, the full TWR is (1.10 × 0.95) − 1, or 4.5%. The arithmetic average is 2.5%, which does not equal the compounded investment growth.

TWR attempts to neutralize external cash-flow timing by measuring what the portfolio earned between flows. It does not ignore investment amounts held within the portfolio; it links percentage returns. The calculation requires reliable valuations at flow dates or a method that estimates the impact. If a contribution occurs during a volatile day and the portfolio is not valued around it, measurement precision can be limited.

How money-weighted return works

MWR is the IRR of the investor’s dated cash flows. Contributions are cash outflows from the investor’s perspective, withdrawals and ending value are inflows. The rate is the discount rate that makes the net present value of those flows equal zero. A spreadsheet or financial calculator can solve it, but the analyst must enter correct dates and signs. Unlike TWR, a large contribution before a return period has more influence than a small contribution.

Suppose an investor contributes $100 at the start, the portfolio rises to $200, and the investor adds $800. The account is then worth $1,000. It falls 50% to $500 by the end. The investor put in $900 and ends with $500. TWR links +100% and −50%, producing 0%. MWR is negative because most of the investor’s money arrived just before the decline. The measures differ because the cash-flow timing differed.

The exact IRR in that simplified two-year example solves −100 − 800/(1+r) + 500/(1+r)^2 = 0, using equal annual periods and contributions at year one. The approximate solution is −41.7% per year. The exercise shows why an investor can have a negative dollar-weighted experience while the portfolio’s time-weighted return is zero. With actual dates, use date-aware IRR rather than assuming equal periods.

A second scenario: contribution before the rise

Start with the same $100 initial investment. After the first period, the account doubles to $200 and the client contributes $800. The combined $1,000 then earns 50%, ending at $1,500. The subperiod returns are +100% and +50%, so TWR is (2.00 × 1.50) − 1, or 200%. MWR is also strongly positive, and the large contribution occurred before the best second-period return. Cash-flow timing helped the investor’s result rather than hurting it.

Compare this with an investor who made the $800 contribution after the second-period gain. That client’s ending wealth would include more cash that did not participate in the rally. TWR can remain unchanged because the portfolio’s return path was the same, while MWR changes. This is why a portfolio manager’s return and a client’s realized experience should not be treated as interchangeable.

When TWR is useful

TWR is commonly used to evaluate portfolio management where clients control contributions and withdrawals. By splitting at external flows, it reduces the effect of client timing and supports comparison with a benchmark measured over the same intervals. It can help answer: how did the investment strategy perform while assets were invested? It does not answer whether a particular investor’s deposit timing was favorable.

TWR is only comparable when the portfolios use similar valuation, fee and reinvestment conventions. A gross manager return should not be compared casually with a net client account return. A benchmark may not reflect the client’s taxes or cash needs. TWR also compresses the return path into a linked number and does not show the amount of capital exposed during each interval. Pair the metric with context.

When MWR is useful

MWR is useful when the investor controls cash-flow timing or when the question is the return earned on the dollars actually committed. Private investments, real estate and individual accounts with irregular contributions can be evaluated with IRR. The result reflects the timing and size of each cash flow, so it can change when a contribution occurs earlier or later even if the underlying sequence of investment returns is unchanged.

MWR has limitations. It can imply that interim cash flows are reinvested at the IRR, which may be unrealistic. Some cash-flow patterns can produce multiple IRRs or no meaningful solution. It can be sensitive to estimated ending value and date conventions. If a project has several sign changes in cash flow, do not rely on one calculator answer without checking whether multiple roots exist. Explain the method and its assumptions.

Worked case: the same strategy, different client outcomes

Two clients use the same strategy. Each strategy rises 20% in the first year and falls 10% in the second. Client A invests $100 at the start and makes no further deposits. The portfolio ends at $108, so TWR is (1.20 × 0.90) − 1 = 8%, and MWR is also 4% annualized over two years, approximately, because there are no intervening cash flows. Client B adds a large amount immediately before the second-year decline. The portfolio’s TWR remains 8%, but the client’s MWR falls because more dollars were exposed to the negative period.

The adviser should report both measures when they serve different purposes and explain the difference. Client B should not conclude that the manager’s strategy was necessarily worse solely because the personal IRR is lower. The manager should not present the TWR as if it were the client’s dollar-weighted experience. Timing and control of cash flows matter to the interpretation.

Worked case: large withdrawal before a rally

An investor begins with $1,000. The portfolio loses 20%, leaving $800. The investor then withdraws $700, leaving $100 invested. The remaining $100 earns 50%, ending at $150. The portfolio’s TWR is (0.80 × 1.50) − 1 = 20%. The investor’s actual dollar experience includes an initial loss and a withdrawal made before the rally; MWR incorporates those flows and can differ substantially. TWR tells the manager’s linked return, not the investor’s total cash outcome.

This example also shows why an ending balance alone is misleading. The account has $150, but the investor previously withdrew $700. A report that ignores the withdrawal would make the investment look disastrous, while a TWR of 20% may omit the client’s experience of the initial loss. The right metric depends on whether the question concerns strategy performance or the return on the investor’s money.

A practical calculation checklist

List each contribution and withdrawal with its date and sign. Record portfolio value at the start, at each flow boundary, and at the end. For TWR, calculate subperiod returns and link them geometrically. For MWR, solve the dated cash-flow equation. Include dividends and fees according to the stated convention. If an ending valuation is estimated, disclose that assumption. Check that all cash flows are from the same investor perspective.

When comparing a portfolio to a benchmark, TWR is often more appropriate because the benchmark does not receive the client’s personal deposits. When evaluating how a client’s capital performed, MWR may be more informative. If the investor controls timing, report the effect rather than implying the manager chose it. When the numbers diverge, explain which cash flow drove the difference.

Read the cash-flow timeline before calculating

A return problem is easier when the timeline is drawn before a formula is selected. Put the opening market value at the left, then mark each external deposit or withdrawal and the portfolio value at that moment. End with the closing value. A deposit is an outflow from the investor’s perspective but an inflow to the account; writing the perspective beside the timeline prevents a sign error. Dividends and interest retained in the portfolio are investment income, not external contributions.

Consider an account worth $10,000 at the start of a quarter. It earns 4%, then the client deposits $2,000, and the remaining quarter earns 1%. If the first subperiod value is $10,400 and the value just after the deposit is $12,400, the closing value after the second return is $12,524. TWR is (1.04 × 1.01) − 1 = 5.04%. The ending account value alone cannot provide that percentage because it includes the added principal.

The dollar-weighted result uses the same timeline differently. The investor’s cash flows are −$10,000 at the start, −$2,000 when the new contribution is made, and +$12,524 at the end. The IRR equation discounts each flow according to its date. It will be close to, but not exactly, 5.04% for the quarter because the extra $2,000 was invested only for the second subperiod. If the dates are irregular, a dated IRR method is preferable to treating every interval as equal.

Separate external flows from portfolio activity

The classification of a cash movement can change both calculations. A client’s deposit, withdrawal, or transfer into or out of the measured portfolio is external. A dividend paid by a security and retained as cash is part of portfolio performance. If the dividend is immediately reinvested, it is still investment activity within the account. A fee deducted from the account is generally part of the client’s net result, although a report may also show a gross return under a stated convention.

Suppose an account earns a 3% market return but pays a $30 advisory fee from a $10,000 balance. A gross performance report and a client’s net account return answer different questions. The analyst should apply the fee convention consistently at each interval and avoid comparing a gross TWR with a net IRR without explaining the difference. Tax payments made from outside the account are another cash-flow question; they do not automatically become portfolio losses simply because the client paid them.

Use the right conclusion when results diverge

A sharp gap between TWR and MWR is a prompt to inspect the timeline, not proof of poor management or poor client judgment. If a client made a large contribution before a strong period, MWR may exceed TWR. If the same contribution came before a decline, MWR may trail it. The portfolio return path can be identical in both cases. The difference measures the interaction between that path and the investor’s dollars at risk over time.

In an advisory review, explain three things in plain language: which measure is being shown, whether it is gross or net of fees, and which deposits or withdrawals materially affected the investor-level figure. A benchmark comparison normally needs a consistent period and return convention. It should not be used to claim that the client earned the benchmark return if personal flows, taxes, or account restrictions changed the result. This distinction protects the usefulness of the metric without disguising a weak investment outcome.

Common mistakes

Do not call TWR and MWR interchangeable. Do not use an arithmetic average of subperiod returns as a linked TWR. Do not leave client deposits out of an IRR. Do not compare the client’s MWR with a benchmark return and attribute all difference to manager skill. Finally, do not present an IRR as a guaranteed reinvestment rate or a forecast.