Your return is not your profit divided by what you paid in. That sum only works if you invested once at the start and never touched it again. Add money along the way, as almost everyone does, and it understates the result.
The reason is simple: money you only added in July has not had a full year to do anything. You have to weigh not just how much you contributed but how long it was there.
Why the simple method misleads
DateEventAmount1 JanuaryStarting balance€10,0001 JulyContribution€5,00031 DecemberClosing value€16,200
The obvious sum: €15,000 in, €16,200 out, so €1,200 on €15,000, or 8%.
But that €5,000 was only present for half the year. Weight each amount by how long it was there and the average capital at work is roughly €12,500, not €15,000. The same €1,200 on €12,500 is 9.6%.
Two honest answers
Time-weighted return measures how the investments performed, ignoring when you put money in. It is what you compare against an index.
Money-weighted return measures how you did, including your timing. It is what your bank balance experienced.
Suppose a fund falls 50% and then doubles. Over the period the fund is flat, so its time-weighted return is 0%. If you invested a little at the start and a lot at the bottom, your money-weighted return is strongly positive. The fund went nowhere. You did well. Both are true.
You want to knowUseHow did this investment perform?Time-weightedWhat did my money earn?Money-weightedHow do I compare to a benchmark?Time-weightedI contribute monthly, how am I doing?Money-weighted
How to calculate the money-weighted figure
It is an internal rate of return, known in Excel as XIRR. List your cashflows with their dates:
- Buys are money flowing in, entered as negative.
- Sells and dividends flow back to you, entered as positive.
- The value at the start counts as money in, the value at the end as money out.
- Solve for the annual rate that makes those flows net to zero.
For the example above that gives 9.6%. One limitation: the equation only has a solution when there is at least one inflow and one outflow. Where cashflows never change sign, a dash is more honest than a confidently wrong number.
Where it still goes wrong by hand
- Forgetting dividends. Income paid out is return, even if it sat in cash rather than being reinvested.
- Leaving out costs. Transaction, currency and service charges reduce your real return and rarely appear alongside your prices.
- Currency. A US share up 10% in dollars might be up 4% or 16% in euros. Every purchase needs converting at the rate of that moment.
- Splits, mergers and reinvestments. A share split changes your quantity without making you better off. One moment of inattention skews an entire spreadsheet history.
This is the real reason most spreadsheets become unreliable after a year or two. Not the formula, but the record-keeping behind it.
Automatically, from your own transactions
Gylder computes the money-weighted return from your actual cashflows using the XIRR convention, shown both in total and per year, alongside what you contributed net. The time-weighted series is used separately for benchmark comparisons and for the risk measures such as Sharpe ratio and volatility.
Two things to know when reading it. The selected period drives every figure on the screen, so a one-year return next to a five-year one is not a like-for-like comparison. And manually added assets have no cashflow ledger, only the valuations you typed, so they are excluded from the contribution-based return rather than given invented flows.
To understand which positions produced the result, read their contribution in percentage points rather than their own return. A small position can double without moving your portfolio much.
Read on
- Sharpe ratio: return set against risk
- Dividend yield: why a rising percentage is rarely good news
- Tracking your portfolio without a spreadsheet