The manager reports what the portfolio earned; the internal rate of return reports what the scheme's money earned, and the difference belongs to whoever chose the flow dates.
Because a time-weighted return gives every year an equal vote and a money-weighted return weights each year by the capital exposed to it. On the Marchwood mandate the two differ by 0.6739 point a year — 4.8012 per cent net against 4.1272 per cent net. The scheme's largest balance met its worst year and its smallest balance met its best. In money the timing cost 6,357,745.18, which is 12.1 times the manager's net outperformance of 0.0556 point a year.
Worked in full in Asset Management by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
A time-weighted return gives every year an equal vote. It strips out money arriving and leaving on dates the manager did not choose and could not refuse, which is why it is the right thing for a manager to report. A money-weighted return — the internal rate of return on the dated cash — gives each year a vote proportional to the capital exposed to it. On the Marchwood mandate the first is 4.8012 per cent a year net and the second is 4.1272 per cent a year net. Both are correct, and neither is an error.
The two disagree exactly to the extent that capital was badly distributed across the years. The scheme was funded with 500,000,000, took in a bulk transfer of 120,000,000 at the start of Year 2, and paid a buy-in premium of 180,000,000 out at the start of Year 3.
| Year | Opening balance | Net return | Gain or loss |
|---|---|---|---|
| Year 1 | 500,000,000.00 | 10.5431% | 52,715,380.00 |
| Year 2 | 672,715,380.00 | -6.1938% | -41,666,554.15 |
| Year 3 | 451,048,825.85 | 12.2431% | 55,222,547.41 |
| Year 4 | 506,271,373.26 | 5.5923% | 28,312,385.52 |
| Year 5 | 534,583,758.78 | 2.8667% | 15,325,120.77 |
Read the first two numeric columns together. The largest balance of the five years, 672,715,380.00, met the worst year, in which the portfolio lost 5.9320 per cent gross and 6.1938 net. The smallest balance, 451,048,825.85, met the best year, in which it made 12.5490 gross and 12.2431 net. That is the whole of the 0.6739 point. The transfer arrived in time for the loss and the premium left in time to miss the recovery.
A percentage gap of this kind is easy to nod at and hard to feel, so convert it. Take the scheme's net capital of 440,000,000, place all of it at inception, leave it alone, and let it earn exactly the portfolio returns it did earn. No skill is added, no timing is improved, and nothing about the manager changes.
| Capital schedule | Terminal value |
|---|---|
| As it actually happened | 549,908,879.55 |
| All at inception, same returns | 556,266,624.73 |
| Cost of the flow timing | 6,357,745.18 |
That 6,357,745.18 is 1.4449 per cent of the capital the scheme put in and 5.78 per cent of the 109,908,879.55 the mandate actually gained. Set it beside total fees of 7,616,440.40 and it is 83.5 per cent of five years of management charges. The fee was negotiated hard and disclosed fully; most of that amount again was lost to a matter on which nobody put a figure.
The vivid version of this story blames the withdrawal, and the arithmetic does not support it. Split the 6,357,745.18 by flow. The 120,000,000 arrived a year late and so missed Year 1's 10.5431 per cent net, worth 14,469,310.14 carried forward to the end. The 180,000,000 that left had earned a cumulative 3.6963 per cent net over Years 1 and 2, which the all-at-inception schedule never has, and that runs 8,111,557.49 the other way. The expensive decision was the late arrival, not the exit.
Two limits, and stating them protects the figure from the first person who wants it dropped.
Nor does the 0.6739 overlap with anything already reported. It does not double-count the fee drag of 0.2875 point a year or the gross outperformance of 0.3432. Those are properties of the portfolio's returns; this is a property of the capital exposed to them.
Beneath the time-weighted return since inception, net, put the money-weighted return since inception, net, on the same basis and to the same number of decimals, and beside them the difference in money. On this mandate it reads 4.8012, 4.1272 and 6,357,745.18. It needs five dated amounts and one closing balance, all of which sit in the administrator's records before the quarter closes, and a single spreadsheet function. There is no system to buy.
The covering note should say what the line is not. It is not an accusation. The flows were sound decisions taken for sound reasons, and neither was a market call. The difference will sometimes be positive, and in those years the line reports a windfall. What changes is the shape of the next conversation: with five years of this line in the pack, the date on which a large transfer is funded stops being a purely administrative output and becomes a choice with a measurable spread.
It is the internal rate of return on the dated cash: 500,000,000 out on day one, 120,000,000 more at the start of Year 2, 180,000,000 back at the start of Year 3, and a closing balance of 549,908,879.55. The rate that balances that sequence is 4.1272 per cent a year, against a time-weighted 4.8012.
No. The flows were the trustees' decisions, taken on their timetable, and the whole point of time-weighted reporting is to keep a manager from lobbying against contributions. The 0.6739 point a year belongs to whoever chose the dates, not to the manager, whose net outperformance was 0.0556 point a year.
No. It is lower when the largest balances meet the weakest years, which is what happened here: 672,715,380.00 met a year that lost 6.1938 per cent net. When flows land well the money-weighted figure exceeds the time-weighted one and the same line reports a gain rather than a cost of 6,357,745.18.
This article is one calculation from Asset Management. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: corporate finance, valuation and markets → · All 1706 articles →
If this book helped — or didn’t — a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.