IRR calculations wrong within Q 2018 upgrade -

Rsta

I have some funds that I purchased about 9 months ago and Quicken 2018 is calculated Avg Annual Return for 1,3,and 5 years as all the same. The one year AAR is correct. I pay for an update "membership" and all I get is a cosmetic gloss over and mistakes in a core function.

NotACPA

Neither the 1 year, 3 years or 5 years term have passed since your purchase.  AAR can't be calculated until the stated term is completed.

Rsta

NotACPA - QW HBRP 2019 said:

Neither the 1 year, 3 years or 5 years term have passed since your purchase.  AAR can't be calculated until the stated term is completed.

Duh.  It should put "N/A" as it did on the 2017 version, not fill in those columns with the less than 1 year calculation.

ColinGruchy

"If it ain't broke, don't fix it".  I wonder if when separating from Intuit that Quicken had to rebuild the software from scratch due to proprietary reasons and in doing so has made a ton of errors in it's haste to get it done?  Hopefully everything is fixed when my 2017 subscription expires in September 2018.

Unknown

ColinGruchy said:

"If it ain't broke, don't fix it".  I wonder if when separating from Intuit that Quicken had to rebuild the software from scratch due to proprietary reasons and in doing so has made a ton of errors in it's haste to get it done?  Hopefully everything is fixed when my 2017 subscription expires in September 2018.

Not even close.  Of course they have the source code, that is the whole point of the sale.  In fact basically the whole Quicken department was sold, change to the new company.  It would have taken years to rewrite Quicken, and would have no advantage over any other company.

Personally I see some of these problems, as the new company trying to be "more responsive", with a mix of the old "we know better than you because we listen to all customers, and then decide, and once we decide we aren't going to change" that I saw for many years.

People have been complaining for years about seeing N/A in these different fields.
And saying "surely you can calculate the values even if I haven't held if for that period of time."  Yes personally I think they should ignore such requests, but if the customers complain enough they take a "stab" at it.  And guess what folks, the people doing this are developers, not financial experts.  If someone suggest that you can do the calculation and ignore the time range, they may very well just code it up.

ColinGruchy

ColinGruchy said:

"If it ain't broke, don't fix it".  I wonder if when separating from Intuit that Quicken had to rebuild the software from scratch due to proprietary reasons and in doing so has made a ton of errors in it's haste to get it done?  Hopefully everything is fixed when my 2017 subscription expires in September 2018.

Perhaps they need a new process improvement model.  Run the requests by financial experts (in-house consultants or external) using a certified requirements management methodology and then manage the change with certified project management methodology.  Don't just code something because someone asked for it.  And do rigorous QA testing before putting it into production.

Unknown

ColinGruchy said:

"If it ain't broke, don't fix it".  I wonder if when separating from Intuit that Quicken had to rebuild the software from scratch due to proprietary reasons and in doing so has made a ton of errors in it's haste to get it done?  Hopefully everything is fixed when my 2017 subscription expires in September 2018.

Perhaps.  I doubt will happen though.  Not that they might get a new "slogan" from time to time, and call that process improvement.

Rsta

ColinGruchy said:

"If it ain't broke, don't fix it".  I wonder if when separating from Intuit that Quicken had to rebuild the software from scratch due to proprietary reasons and in doing so has made a ton of errors in it's haste to get it done?  Hopefully everything is fixed when my 2017 subscription expires in September 2018.

Well I wouldn't care if they reported return based on investment performance before I owned it.  But to calculate the return from the less than 1 year history and report it as 1,3, and 5 year ARR, that's just incompetent.

ColinGruchy

ColinGruchy said:

"If it ain't broke, don't fix it".  I wonder if when separating from Intuit that Quicken had to rebuild the software from scratch due to proprietary reasons and in doing so has made a ton of errors in it's haste to get it done?  Hopefully everything is fixed when my 2017 subscription expires in September 2018.

Does it make the aggregate reporting for your whole portfolio inaccurate?  For example does it change the 5 year ARR for your entire portfolio?  I had to delete accounts from Investment Portfolio view because they were showing negative 99% return after transferring securities from one account to another.  This was impacting my overall portfolio returns.

Rsta

ColinGruchy said:

"If it ain't broke, don't fix it".  I wonder if when separating from Intuit that Quicken had to rebuild the software from scratch due to proprietary reasons and in doing so has made a ton of errors in it's haste to get it done?  Hopefully everything is fixed when my 2017 subscription expires in September 2018.

Honestly I have no idea. I have a few accounts with 20-40 investments.  Maybe I'll pick a small one and try it.

Dennis Nelson

Having problem with investment performance report calculating a return on Amazon stock bought in July 2017.  Version R14.1 Build 25.1.14.9.  It properly calculates return in 2017 but uses a zero opening value in 2018.  All other securities I hold are having their performance properly calculated.

Ok, just solved problem by deleting all transactions under that security, deleting the security and then repurchasing the security on the original date.  Don't understand what was wrong with the other security but that fixed it.

Rsta

That seems pretty extreme. I'm not deleting and re-entering all transactions around sticks I've held for years. So you are saying the 3 and 5 year returns are correct now?

Jim_Harman

Dennis, perhaps your original situation was that somehow a share price of zero got into the price history on or just before 1/1/2018. You might want to check that before deleting and re-entering everything if you see this problem again.

Rsta, if you have held the securities for less than the analysis period or if the analysis period is less than one year (YTD for example) the average annual return (IRR) can be calculated but it will give confusing results. This is especially true for securities that have varying share prices. In previous versions of Quicken, it would display N/A in the Investing/Portfolio views in this case. In QWin 2018, they chose to display the correctly calculated but confusing number instead.

To get a better feel for what is going on, I suggest you experiment with the Investment Performance report (Reports/Investing/Investment Performance) which shows the balances and cash flows that contribute to the IRR calculations.

I think you will find that if you plug the same values into Excel's XIRR function, you will get identical or nearly idnentical results.

Rsta

Jim Harman said:

Dennis, perhaps your original situation was that somehow a share price of zero got into the price history on or just before 1/1/2018. You might want to check that before deleting and re-entering everything if you see this problem again.

Rsta, if you have held the securities for less than the analysis period or if the analysis period is less than one year (YTD for example) the average annual return (IRR) can be calculated but it will give confusing results. This is especially true for securities that have varying share prices. In previous versions of Quicken, it would display N/A in the Investing/Portfolio views in this case. In QWin 2018, they chose to display the correctly calculated but confusing number instead.

To get a better feel for what is going on, I suggest you experiment with the Investment Performance report (Reports/Investing/Investment Performance) which shows the balances and cash flows that contribute to the IRR calculations.

I think you will find that if you plug the same values into Excel's XIRR function, you will get identical or nearly idnentical results.

Thanks I'll check that out. Personally for something I've held for less than a year, I'd rather see N/A than some confusing projection that is the same number for 1,3, and 5 year IRRs - which can't be correct. If I am sorting 5 year IRR, I don't want to see some made up number. What I'd really like to see is 10 year IRR - maybe I can get that from the Investment Performance report - let me check.  Thanks!

Jim_Harman

Based on some experiments today, I now see that there are some differences between Excel's XIRR function compared to Quicken's average annual return calculations in the Investment performance report. For my experiments I had just starting and ending values with no deposits or withdrawals during the period. This is like a stock that you held for the entire period with no buys, sells, or distributions.

I also assumed that the share price does not vary for a few days before and after the start and end of the period, so I could see the impact of adjusting the dates without affecting the prices.

Findings:

1) If the balance is nonzero at the start and end of the period, the answers are identical but you need to add one day to the period in Excel, so if the date range in Quicken is "Current year", the report header says 1/1 - 12/31 but you have to set the dates in Excel's XIRR function to 12/31 - 12/31. 

2) Quicken uses the share price on the day before the start of the date range to compute the starting balance or "Beg Mkt Value" in the report, This makes sense, because that would normally be the closing price for that day.

3) If the balance is zero at the start of the period, i.e. the initial cash flow is zero, Excel sets the IRR to 0. Quicken starts its calculation on the initial purchase date and ends it on the final sale date, so it is showing the annualized return for the time you owned the security. This is why the 1-year, 3-year, and 5-year numbers are all the same for securities that have been purchased recently. I think in previous versions Quicken showed N/A if the initial balance was zero.

4) If the initial cash flow in Excel is close to zero and the subsequent ones are much larger, Excel's results match Quicken's - it shows the annualized return for the period you owned the investment.

5) If you set the ending date in the future, Quicken assumes that the ending value will be the same as today's value. This is as good an assumption as any. 

6) Both Quicken and Excel assume years are 365 days, so a 20% gain in a leap year shows as 19.94%.

There may be other anomalies; I will report back if I see them.

Conclusion:

The IRR is useful, but not for every situation, especially as I said for analysis periods of less than one year or holding periods less than the analysis period.

I agree it would be less confusing if Quicken went back to showing N/A in the Investing/Performance views for situations when the IRR is not meaningful, i.e. the value is zero at the start of the period.