Is AAR useless when looking at investment performance?

Alex100

My apologies for the somewhat clickbait-y title of this question, but I've never understood Quicken's fascination with reporting AAR in its investment reports and am hoping someone can explain to me why it's useful.  When I look at my investment portfolios, one of the primary questions I want answered is performance: how are my investments performing year-to-date, quarter-to-date, or some arbitrary date range.  I think the metric that best answers that question is ROI-percentage, and the only place I've found in Quicken that reports ROI is the Investments "portfolio" view.  All the reports (specifically the investment performance report) show AAR.  

I understand the math behind AAR (and IRR); what I don't understand is why anyone would want to view or measure their investments that way.  If I login to ETrade to look at my account performance, it doesn't show me AAR for YTD performance, it shows ROI. If it showed me AAR, my performance 2 months into the year would be like 82% which is nonsensical, instead it shows me ROI which is more like 4.5%.  (I'm making these numbers up, but you get the point) 

Investment advisers don't use AAR when talking to their clients, people don't use AAR when talking about investments with each other, so why does Quicken insist on using it?  What am I missing?  Thanks for any feedback!

Jim_Harman

You are right that IRR (AKA AAR) is not useful for periods of less than one year, unless you are looking at a fixed price investment like a money market fund. 

However for multi year performance it is more useful than other measures, because it takes compounding into account. 

It is also the measure that mutual funds use to show their long term performance. 

One way to use the ARR for a YTD report is to set the end date of the report to 12/31.

Alex100

Thanks Jim, that makes sense - so for an account with a mix of investments (stocks, equity mutual funds, bonds) is AAR still a reasonable way to look at performance, assuming the time period is greater than 1 year?  

Also, and I guess this is a related question, if I had an account with only non-dividend paying stocks, would AAR look about the same as ROI since there is no compounding to take into consideration?  

Jim_Harman

If there are no transactions during the year, I think the 12 month ROI and IRR will be the same.  The ROI is not annualized, so for any other period ,the results will be different. 

Tom Young

"You are right that IRR (AKA AAR) is not useful for periods of less than one year, unless you are looking at a fixed price investment like a money market fund. "

I completely disagree.  There's no "magic" about "a year" that somehow makes the IRR a nonsensical number for shorter periods.  The IRR is what it is: a percentage figure, typically expressed as an "annual rate", though other periods could be used, of a security's or an account's performance and it's perfectly valid to compare the number to other securities or accounts or "the market" for the same period.

If I have a stock that's gone up by 50% in half a year, resulting in an IRR of something on the order of 126%, where "the market" has gone up by 25% in the same period for an IRR of around 57%, which one has performed better over that time?  The IRR here is purely a backward-looking metric and provides a very clear picture of relative performance.

Conversely, if I have a stock that's gone up 50% in 364 days, for an IRR of about 50%, (you can figure out that one in your head), how is that not useful?  Barring some disaster or earth-shaking even on the 365th day, I know the IRR for the year is going to be 50%.  Was that a good investment, or not?  I can't say "yes" or "no" on day 364 but I can on day 365?

Jim_Harman

@Tom Young, perhaps I misspoke when I said  IRR is not useful for periods of less than one year.

Really you just need to understand what it is telling you. For a period of less than one year, an IRR calculation assumes that the performance will continue at the same rate for the rest of the year. Is that a good assumption? It depends on the investment. 

Also published investment performance data typically shows raw percentage changes for quarterly and YTD periods matching Quicken's ROI data,  but annualized numbers matching the IRR for annual and longer periods.

Tom Young

"For a period of less than one year, an IRR calculation assumes that the performance will continue at the same rate for the rest of the year."

Yes, that's the "annual rate" aspect of the calculation and I think that's what befuddles people.  But as a measure of performance it's perfectly valid.  And even pure backwards-looking calculations assume that cash flows can actually be invested at the IRR rate, which probably also isn't accurate.  It's no a "perfect" metric; the better metric would be to use your own cost of capital, but nobody, really, knows that.

Alex100

Tom Young said:

Yes, that's the "annual rate" aspect of the calculation and I think that's what befuddles people.  But as a measure of performance it's perfectly valid. 

Except that typically people compare their performance against an index.  If the S&P is up 4.8% so far this year and I said, "that's nothing, my portfolio IRR is actually up 78%" I'd be labeled a fool. I'd also be a fool to assume that I'm going to get the same return for the next 9 months.  That's why Quicken's Investment performance reports are basically useless for me, unless I'm looking at a long time period I guess.  I'm not suggesting the number is not mathematically correct, I'm saying it's not helpful, and I can't for the life of me understand why those reports don't at least also include ROI.  

cranky_user

Can't you use the Investment Performance report with date range set to Yearly and Current Year?  That's what I use.

I agree Quicken is very confusing.  I never use the Investing Portfolio tab anymore.  The ROI calculation there is useless if you are buying/selling shares throughout the year as its calculation is just the change in share price * current share balance.

Jim_Harman

Yes, setting the date range to yearly and current year does produce a useful YTD value in the investing performance report. The underlying assumption is that the return will be flat for the rest of the year. This may or may not be valid for your investment. 

I think if you take a closer look at the ROI(%) calculation, you will find that it does include purchase and sales. 

Alex100

Wow, that was the trick - thank you both!  Setting the date range to yearly seems to be the magic bullet that makes the AAR not "annualize" the gain/loss.  I can't believe I've been a Quicken user for this many years and have not been able to get those reports to do what I want.