Lifetime Planner IDEA: Set Rate of Return Based on Investment Portfolio Weighted Asset Classes

Scooterlam

Lifetime Planner IDEA: Set Rate of Return Based on Investment Portfolio Weighted Asset Classes

SUMMARY:

Users often struggle with setting investment rate of return (RoR) in their retirement planning tools, such as lifetime planner (LTP). The investment rate of return is generally, but not always, tied to the user's investment portfolio's asset allocation (stocks, bonds, cash et al) Yet, many look to various historic market index returns, such as DOW, S&P and NASDAQ alone to guide their LTP RoR assumption. Some have taken a reasoned approach to setting future returns while others just ask "some guy on the internet" what they should use….

This proposed, optional LTP feature, allows for a calculated weighted rate of return based upon the user's Quicken investment portfolio's asset allocation. Reference 1 / Image 1.

The purpose is to ensure the user has some RoR basis in their assumption and avoid wildly different return assumptions than what their actual portfolio investments have historically returned.

So….The summation of the investment portfolio's individual asset class weighting times the historical return for that asset class results in a weighted portfolio return.

Please vote below!   Add your comments and thoughts.

Image 1 - LTP Weighted RoR based on User's Actual Investment Portfolio

PROPOSED FUNCTIONALITY:

The overall workflow allows the user to optionally base their LTP RoR on existing functionality (image 2) or use their actual investment portfolio to derive a weighted, asset class based RoR. Image 1 / 2.

The New Workflow Overlay on Existing Return Dialog (Image 2):

1. adds a checkbox that performs this calculation in the existing return dialog
2. checks for errors and guides the user to resolution / next steps. ex. no user portfolio
3. allows user validation of the new calculated weighted RoR
4. upon clicking done, LTP uses these newly populated, portfolio weighted RoR figures in the yearly portfolio cashflow calculations, as normal.

Image 2: Existing Functionality + Weighted RoR New Basic Workflow

THE CHANGE:

• Much of the data structure is already available. Still, some needs to be built… Image 4

• The user must have their investment portfolio defined in the investing>portfolio view and ticker asset classes correctly assigned / downloaded. Image 5

• If the user does not use Quicken investments portfolio feature, then offer them the abiliity to enter the "target" portfolio asset class from which to base their weighted return. Image 4.

• Ensure that historical asset class data is available and updated timely.

• Note that some historical data tables are already in use in order to calculate "Expected Return" and "Expected Risk" fields in the Investing>Asset Allocation dialog. Source, timeframe and quality unknown. It returns strange figures. Image 4.

• Ensure the ability to run RoR weighting against existing taxable, tax-deferred and spousal accounts, as shown in the original return dialog. Image 2.

• Dual purpose the "Actual" allocation table and chart as the "Before Retirement" allocation table and chart, as required in LTP's return assumption dialog. Image 4.

• Build an "After Retirement" asset allocation table and chart, similar to the Target allocation. This or something equivalent would be needed to define the post-retirement allocation required in LTP's Return assumption dialog. Image 4.

• After initial build, consider expanding on the 6 original asset classes to include custom classes along with the ability for user defined RoR for such custom classes.

• Additionally, allow the user to enter "self-defined" asset class returns, when they feel historical returns are inappropriate. These self-defined asset class returns could be, for example, forward looking forecasts from major investment firms. Reference 2.

• Allow user to further define time periods (other than pre / post-retirement) with corresponding asset allocations to simulate changing levels of risk as the user ages or life events occur.

IMAGE 4: Investing>Asset Allocation Changes

Image 5 - Security List - Tickers and Associated Asset Classes

REFERENCES:

1. How to Calculate Expected Portfolio Return
2. Lifetime Planner: Resources for Setting Rate of Return and Inflation Assumptions

Please vote below!   Add your comments and thoughts.

Rocket J Squirrel

QWin can show me average annual returns for 1, 3, and 5 years for each security, account, and for my whole portfolio. Can't I just plug in the portfolio's 5 year AAR into this LTP parameter?

Scooterlam

@Rocket J Squirrel I think that 5 year span is too short. IMO, not enough business cycles historically, to project out decades into the future. I prefer to go back decades to extablish an asset class's return track record and project that going forward. Many data sources go back to 1871, depending on the asset class.

Unless I am fundamentally misunderstanding the concept of "mean reversion", I believe that this is a sensible approach. While there is considerable debate about using historical return averages in forecasting, I think that it is one of the best sources of guideance going foward. Past performance…..statement that everyone quotes but few follow. Why? Isn't there a better set of data? Read on….

The reference #2 in the OP is a whole bunch of forward looking guideance from the "experts". A review of last couple of their forecasting years v 2023 forecasts will have you whiplashed….how is this possible? Forecasting is hard business but these were the same experts declaring the death of the 60/40 portfolio, bonds and the "4% rule" in the last couple years…. Its a wonder why normal folk are confused. It doesnt seem like the thoughtful forecasting I would expect.

In terms of the Average Annual Return calculation in Quicken, I am not sure if this is appropriate even if it did go out 10 or 20 years. Is this an arithmatic average or geometric average? I couldn't tell. Geometric averages are otherwise known as "compounded annual growth rate" and take in the volitility of the return over time. They are preferred when measuring return performance of an equity as it takes into account compounding vs. a simple arithmatic mean.

Have a look at Portfolio Visulizer. They have an excellent tool to backtest portfolios based on ticker or assett class. From there you can put in the return assumptions into LTP. A lot more work for sure.

Rocket J Squirrel

@Scooterlam

I think that 5 year span is too short. IMO, not enough business cycles historically, to project out decades into the future. I prefer to go back decades to establish an asset class's return track record and project that going forward. Many data sources go back to 1871, depending on the asset class.

OK, that may be a reasonable approach… or it may not. Going farther back in time might introduce statistics which were true in their day but no longer apply and would now skew the results.

Imagine you held stock in a humble GPU chip designer just 10 years ago. NVDA will never revert to that level, entirely due to serendipity. 🤑 So including those old statistics would skew the results downward inappropriately.

I think in the big picture, there is so much GIGO in LTP that we can't place too much reliance on its prediction. Lacking even something as basic as Monte Carlo simulation, it's a pretty dull knife. That said, I do keep mine as accurately set up as possible just as another point of reference.

Please don't take this as argument. I fully agree that it would be great if LTP were improved. I just don't see it happening, so I look for whatever inputs I can provide it without the research taking up the rest of my Life Time. 😵

Scooterlam

https://community.quicken.com/discussion/comment/20365339#Comment_20365339

Your very welcome to use whatever return and associated time horizon you feel represents your portfolio make-up. You can tie you return to a fantasy idea (GIGO) or you can tie your return to some reasonable basis, such as asset class historic returns, which is the point of the IDEA.

By using the NVIDIA example, you've cherry picked a recent high-flying, high risk, individual tech stock as representative behavior of a well diversified portfolio to try to make your point (historic returns may not always work). Good grief, you can do better!

Rocket J Squirrel

@Scooterlam

By using the NVIDIA example, you've cherry picked a recent high-flying, high risk, individual tech stock as representative behavior of a well diversified portfolio to try to make your point (historic returns may not always work). Good grief, you can do better!

I wasn't suggesting using a single stock to represent an entire portfolio, good grief! It would be one of many in the portfolio. I was simply presenting an example showing that going back too far in time, you risk incorporating history which will never repeat itself and could skew the results.