@wallace471 Bob - Let me start by saying yours is a very interesting and informative analysis! Thanks for doing this. Three thoughts on what you've observed:
1 - I suspect PRC survey users answered the questions from a 20-30 year perspective since PRC is designed to project a full retirement timeframe, not just 10 yrs out. Some may configure it for just 10 yrs., but I think most go to end of life, so 20-30 yrs much more likely.
2 - Bias - One thing about PRC or most planning tools is you can construct as rosy an outcome as you wish by upping your inputs. Coming off ten years of above-average returns, I suspect some users responded with a bias towards a much more positive outcome than is realistic.
3 - I also suspect some users simply didn't think REAL ROR (or pay attention to the instructions to enter real ROR and not nominal) so some of the higher ROR responses may reflect nominal vs real, especially since most published reports express returns as nominal and not real.
Having said the above, it's revealing (in a somewhat troubling way) that even the most popular responses (by asset class) are outside the ranges Vanguard proposes, and some by a large margin. Begs the question as to how many users are vastly over or underestimating their results and consequently making life decisions based on erroneous projections.
I wonder, therefore, if the new web-based tool should incorporate warning popup messages if a number or percentage is entered that is way off from the norm. The user can still enter whatever they want but it would prompt them to recheck as a safeguard.
Yes Bob thanks for doing this ????. My take away (as if that's counts for anything????) is PRC Forum users are one smart group! I would believe them before Vanguard or any of the other "professionals". Remember, there is not such thing as a professional financial predictor. Forum users have just as big of brains and have access to all kinds of information on the web if they take the time to do their own research, so that they can come to their own valid conclusions. Gold star to Forum users ????. Incorporating warning popup messages may be a good idea, but how do you define the "norm". Predictions from investment houses, historical averages, Farmers Almanac... not sure myself. Not trying to be a wise a$$, just trying to be useful ????. Thoughts from other lurking ???? Forum users??????? I know you're out there.
@pizzaman it would be like on some sites where if you enter a number that doesn't fall within a certain range (or is misformatted) it throws up a warning that says something like "the value entered isn't typical - please check and re-enter"
yes, someone would have to define what an acceptable range is for certain cells. It can be broad enough to let most rational entries pass, but its mostly to prevent someone from entering 23% as an equity ROR when they intended to enter 2.3%. A way to keep people within the guardrails or reality. Stuff like that.
You could also allow the user to disable data entry warnings up front
Yea that sounds good! Maybe also add an extra warning about Real vs Nominal %.
@pizzaman Agreed. Gives users a sense that someone's looking out for them as they formulate their plan. Some will like it, others may disable it. Their choice.
Regarding the 2023 survey, it would be nice if we had updated information regarding everyone's input parameters and Pralana online configurations now that the online version is in place. Is there any interest in updating the survey from early 2023? This would give all of us a good start to 2025, especially to id realistic results. I learned a wealth of information from the previous survey, but I am not the person to coordinate such a task.
Is it possible that Pralana itself can summarize key input data/assumptions (with privacy in mind) and then interested members can comment?
I would love to see updated survey results related to ROR and inflation, but this thread has been helpful to me. Before seeing this I posted a question on EarlyRetirement forum and the suggestions went from "just use historical" to "make it equal to inflation". I am past the "will I be OK" point. I am trying to optimize and want to use good numbers. On a related note, I too am struggling with the real vs nominal. A suggestion would be for the Online version to simply display the calculation of whichever one you do not enter. If you enter nominal, show the calculated real return. Might even be helpful to show some historical data on this screen and give user ability to choose it with a modifier up or down.
- @patton525 I launched the survey back in 2023, but to your point, it would be much easier if PRC could harvest the de-identified raw data and show it graphically so users could refer to it as they build their own model.
I would love to see a table of historical returns for items other than the built-in asset classes. For example, small-cap value, large-cap growth, 10+ year US Treasuries, Gold, & managed futures. I'm working on this now and will soon post my data for others to use if the choose to do so.
Here is a table with some example information from various providers. You can google each provider to get the full report. They typically call it Capital Market Assumptions. I don't show the standard deviations but most providers do provide that as well. Many providers also provide a range of possible returns in addition to the point estimates. The estimates are as of the 4th quarter of 2024. For equities, although there is a large difference between providers, they are well below historical averages and you can read the reports to get the explanations. For bonds, the range is much less due to the fact that current YTM is a good proxy for future returns for any given duration. Research Affiliates has a great website with tons of this type of data and easy to understand assumptions along with historical data to easily compare history to future expectations. The interface is great but you have to buy-in to their approach. They claim that since they have been doing this that their estimates have been reasonable matches to subsequent returns.
| Nominal ROR For Common Asset Classes | ||||||||||
| Firm | Period | Global Equities X-US | US Equities | US Small Cap | US Value | REITS | US Bond Agg | US Treasury | US TIPS | Inflation |
| J.P. Morgan | 10-15-Yrs | 7.1% | 6.7% | 6.9% | 7.7% | 8.0% | 4.6% | 3.8% | 4.1% | 2.4% |
| Black Rock | 10-Yrs | 8.0% | 6.2% | 5.4% | 6.9% | 3.7% | 3.4% | 4.0% | 2.4% | |
| Fidelity | 20-Yrs | 6.8% | 5.7% | 5.2% | 4.8% | 5.1% | 2.6% | |||
| Invesco | 10-Yrs | 6.6% | 4.7% | 7.7% | 6.2% | 4.5% | 3.9% | 4.3% | 2.1% | |
| Voya | 10-Yrs | 4.7% | 4.1% | 4.5% | 3.7% | 2.9% | 2.4% | |||
| Vanguard | 10-Yrs | 7.9% | 3.8% | 5.2% | 4.8% | 4.8% | 4.6% | 3.9% | 2.4% | |
| Research Affilites | 10-YRs | 6.8% | 3.2% | 6.8% | 3.9% | 6.5% | 4.9% | 4.8% | 2.5% | |
| Goldman Sachs | 10-Yrs | 3.0% | 2.4% | |||||||
| High | 8.0% | 6.7% | 7.7% | 7.7% | 8.0% | 5.2% | 4.8% | 5.1% | 2.6% | |
| Low | 6.6% | 3.0% | 4.1% | 3.9% | 4.8% | 3.7% | 3.4% | 2.9% | 2.1% | |
| Average | 7.2% | 4.8% | 6.2% | 5.3% | 6.5% | 4.5% | 4.2% | 4.1% | 2.4% | |
| Median | 7.0% | 4.7% | 6.8% | 4.9% | 6.5% | 4.6% | 4.3% | 4.1% | 2.4% | |
| Real ROR For Common Asset Classes | ||||||||||
| Firm | Period | Global Equities X-US | US Equities | US Small Cap | US Value | REITS | US Bond Agg | US Treasury | US TIPS | Inflation |
| J.P. Morgan | 10-15-Yrs | 4.7% | 4.3% | 4.5% | 5.3% | 5.6% | 2.2% | 1.4% | 1.7% | 2.4% |
| Black Rock | 10-Yrs | 5.6% | 3.8% | 3.0% | 4.5% | 1.3% | 1.0% | 1.6% | 2.4% | |
| Fidelity | 20-Yrs | 4.2% | 3.1% | 2.6% | 2.2% | 2.5% | 2.6% | |||
| Invesco | 10-Yrs | 4.5% | 2.6% | 5.6% | 4.1% | 2.4% | 1.8% | 2.2% | 2.1% | |
| Voya | 10-Yrs | 2.3% | 1.7% | 2.1% | 1.3% | 0.5% | 2.4% | |||
| Vanguard | 10-Yrs | 5.5% | 1.4% | 2.8% | 2.4% | 2.4% | 2.2% | 1.5% | 2.4% | |
| Research Affilites | 10-YRs | 4.3% | 0.7% | 4.3% | 1.4% | 4.0% | 2.4% | 2.3% | 2.5% | |
| Goldman Sachs | 10-Yrs | 0.6% | 2.4% | |||||||
| High | 5.6% | 4.3% | 5.6% | 5.3% | 5.6% | 2.6% | 2.3% | 2.5% | 2.6% | |
| Low | 4.2% | 0.6% | 1.7% | 1.4% | 2.4% | 1.3% | 1.0% | 0.5% | 2.1% | |
| Average | 4.8% | 2.4% | 3.8% | 2.9% | 4.1% | 2.1% | 1.8% | 1.7% | 2.4% | |
| Median | 4.6% | 2.5% | 4.3% | 2.5% | 4.1% | 2.4% | 2.0% | 1.7% | 2.4% | |
@golich428 Thanks for the information. I'm hoping we can create an XLS with every (reasonable) asset class annual nominal returns back to 1928 that all users can access and import into their model for historical analysis. We already have information on the major items but nothing for managed futures, for example. In many cases, the asset class has not been in existence since 1928. I can find data for both large-cap & small-cap stocks but not large-cap GROWTH & small-cap VALUE.
Anyway, I will contribute what I can find in an easy-to-use format for import into Pralana for personalized historical modeling. It will not be posted here but somewhere else where it belongs (online Build section?).
@golich428 Here's an expanded version of the survey method, with 41 firms: https://www.horizonactuarial.com/survey-of-capital-market-assumptions .
I like this because it gives both 10y and 20y summaries of 41 firms, including I think the 7 firms cited by Greg above, but including many more. (It shows each firms estimate with a "dot" on graphs, but deliberately keeps each firm's estimate anonymous to increase validity.) I find the charts at the back more useful than the averages at the front since they give the median (rather than the average) and also shows the "spread" of the estimates so you get a sense of the range for any asset class. (I attached a snip of their 10y sample results.)
@nc-cpl Just to clarify, I didn't post the above table; I gave a link to an actuarial firm who posts the average / median / and percentiles of returns estimates of 41 firms, including those in Greg's table. What I do for myself is adjust their average nominal returns using their estimated inflation, using the equation Real Return=(1+n)/(1+i)-1, where n is nominal return and i is inflation. (You don't simply "subtract" the two since their returns are geometric, ie annualized over the entire periods in question).
For the median estimates among 26 firms here are their nominal medians, with my calculated real. But do note there is a wide range among estimates (see snip I posted in my previous post). For instance, US large 25th to 75th percentile among the estimates ranges from 6.1-7.7 (3.6-5.2 real). I was surprised how far the results differed from the 7 firm survey above, especially US equity!
26 firms, median estimated geometric returns:
| nominal 10y | real 10y | nominal 20y | real 20y | |
| us large | 6.6% | 4.1% | 6.7% | 4.2% |
| us small/mid | 7.0% | 4.5% | 7.3% | 4.8% |
| non-us dev | 7.5% | 5.0% | 7.5% | 5.0% |
| non-us emrg | 7.8% | 5.3% | 8.3% | 5.8% |
| inflation | 2.4% | 2.4% | ||
| CORP core bonds | 4.9% | 2.4% | 4.9% | 2.4% |
| tips | 4.3% | 1.9% | 4.2% | 1.8% |
| cash (us TREAS) | 3.5% | 1.1% | 3.3% | 0.9% |
Just curious, if you invest in Total Market or S&P index funds, would you simply ballpark the ROR based on Large, small emerging?
