Message Font: Serif | Sans-Serif

No. of Recommendations: 0
Has anyone else noticed that Quicken 98 calculates the IRR (internal rate of return, also known as average annual total return) incorrectly in its "Investment Performance report"?

You can verify this with the following example:
1/1/96 buy 1,000 shares of something at \$1 per share
7/2/96 sell all 1,000 shares for \$2 per share

Do the Investment Performance report, and use "customize" to set "subtotal by" to subtotal by security. Use a date range of 1/1/96 to 1/1/97 (or anything 1/1/96 or earlier to 7/2/96 or later).

If you have an annually compounded investement that yields 200%, and you invest \$1,000 on 1/1 (as in the example above) and close out the investment half a year later (7/2 in the example), you would receive \$2,000 back (again, like in the example above). This is because 200% interest on \$1000 is \$2000, but since it was only half a year you'd only get half that, or \$1000. \$1000 interest plus the \$1000 original investment is \$2000. There's no compounding of interest since I've specified annual compounding and it's been less than a year.

Quicken's help text says they use annual compounding, but the answer in the report is roughly 300%, not the correct answer of 200%. 300% would be the correct answer if the interest was compounded twice a year.

I've called Intuit and they admit the report is wrong, but call it a "program limitation". The reason the results are wrong is that the formula they use is improperly derived. Incidentally, Excel's XIRR function uses the same formula and gives the same incorrect answer. Microsoft admits the answer is wrong when you use periods less than a year. In fact, Quicken and Excel are wrong when you use periods that aren't exact multiples of a year.

I noticed that on Intuit's support web page there's the question "Why is the annual return on my Investment Performance report too high?" but their suggestion is to use a report period of one year (not because the answers are wrong, but because over a year they are easier to understand). In my example above, changing the report period makes no difference as long as it includes both the buy and sell dates.