FREE NEWSLETTER
I SAW A GRAPH recently generated by some retirement-planning software. It showed the investor enjoying substantial portfolio growth over the course of his 30-year retirement. Forget running out of money. This particular software program says the guy’ll be a 90-year-old multimillionaire.
My curiosity piqued, I used the same software to run numbers for my finances. I ran optimistic and pessimistic assumptions. I entered my monthly expenses and my fixed income. I tried to run out of money, or at least show a decline, in my total assets. No luck. I just kept getting wealthier over the next 20 years—assuming I have 20 more years.
Even when inflation hits my buying power and I’m compelled to pull more from cash, there’s no negative impact on my assets’ growth. I also projected an extra withdrawal for a new car. Why not? The software still showed me getting richer.
This experiment led me to question the software’s validity, or at least the validity of the assumptions used. This software assumed a 4% withdrawal rate, 4% inflation and an 8% annual rate of return.
Many people put a lot of faith in planning software, betting their financial life on assumptions, projections and past performance. The assumptions are usually theirs, but are they accurate? Do they reflect the reality of how they’ll live in retirement?
For example, all these planning programs factor in inflation. Is your personal inflation rate the same as the Consumer Price Index? Probably not. By adjusting your buying, you can lower your true inflation rate. I could, for instance, lower my thermostat during winter to 55 degrees. Such an action, however, may also reduce my retirement’s fun factor.
Or you can just game the system by fudging your inflation assumption. If you use a 4% inflation rate, does your plan fail? Tweak it to 3.5%, and you might find instant success. This seems risky to me.
If you’re assuming an 8% average annual investment return, does your portfolio hold enough stocks to notch that sort of growth? Do you plan on changing to more conservative investments as you age? And what about that warning that “past performance is no guarantee of future results”?
One person I spoke to is convinced retirement-income needs are overstated, and this is partly fueled by faulty software projections. The projected shortfalls create a demand for investment products and lead folks to over-save, my acquaintance contends. I don’t know about that. But I do know people are making critical long-term bets using assumptions that may prove far too optimistic or conservative.
Thirty years is the typical retirement planning assumption. But U.S. life expectancy for a 55-year-old man is 24 years. At 60, it’s 20 years and, at 65, it’s only 17 years. For women, life expectancy is some three years longer. How many people are likely to be retired for 30 years? It’s not a bad goal. Just understand it might not be realized.
I’ve talked to some people who assume they’ll tap home equity, or maybe sell their home at some point in retirement, to cover their retirement’s costs. But what’s the backup plan if they change their mind?
Are people honest with their expected expenses? Detailed planning, with folks often trying to account for every dollar over many decades, is impossible. My advice: Keep it simple, and use your current expenses plus any obvious changes, such as paying off a mortgage if that’s scheduled to occur.
Research shows that the average retired household’s consumption falls between 0.75% and 0.8% each year. Retirees in good health and with more wealth don’t adjust their spending as sharply as those with poorer health and less money.
Are retirees adjusting their spending because they’re compelled to or does their spending naturally decline? I never planned to spend less and we haven’t, but we certainly spend differently. I just overheard a discussion between Connie and our construction guru son about kitchen cabinets and a new floor. What can they be up to?
Planning for retirement is no easy task, especially when that planning involves income that can be variable, expenses that we can’t fully control and assumptions that are little more than guesses. I see software as one tool, but not the sole driver of decisions. Step back and conduct a sanity check on your plan. Consider having a neutral pair of eyes review your assumptions.
Inflation assumptions should be viewed critically as well. There is reported inflation and lived inflation, and the two are different.
I have no car payment nor a need for a new one, and my mortgage is fixed, so the inflation that hit the used/new car markets and the “owner equivalent rent” that weigh somewhat heavily in USG calculations are meaningless to me. And those are major expenditure categories for most.
I’m still working. A COLA increase of 5% after ’22 resulted in a net gain for me. Other than groceries, we didn’t feel much inflation. We’re in our late 50s; we don’t buy “stuff” – we have too much already. My spouse is retired and I work in a casual office – so clothing budget is very low. Utilities didn’t move much in our area. And while we budget for car and home repairs, we have had very few over the years.
Online complaints about a $5 increase in some streaming service are amusing: cancel it if you don’t want to pay it, but c’mon, it is $60 a year!
Look at your expenses and figure out what may increase and decrease during your retirement. And what is necessary vs what is discretionary.
Assuming 4% inflation – which other than the pandemic induced increases of ’22 we haven’t seen in decades – strikes me as unreal and overly pessimistic.
Interesting Article. As a (now retired) financial planner that has run hundreds of projections for clients on reputable planning software, a few comments:
Retirement planning software is an extremely valuable tool, but must be used properly and is only very important in certain situations.
As long as people’s financial plans have discretionary expenses included, in dire circumstances, if one is disciplined, these can be eliminated. One needs to look at what their core expenses are and determine if catastrophe should occur they could still cover them
8% average future return in a diversified portfolio? Ha! If wishes were fishes we’d all cast nets.
What’s everyone’s favorite retirement planning software? (I use an Excel spreadsheet for forecasting yearly account gains, withdrawals, Roth conversions, taxes, social security, RMD’s etc. but haven’t tried looking for anything fancier.) I do use and recommend Holistiplan for tax planning, though I do wish it was less expensive.
Have used The Complete Retirement Planner for the past three years. It’s a good basic program and it allows you to provide different investment and expense assumptions. The developer is very responsive to questions and gets back right away, usually the same day. It’s spreadsheet based (need Excel 2016 or later) and not flashy like the big names, but it doesn’t need to be for what I use it for, which is to make sure financially I’m in good shape going forward. Another plus is that the data stays on your computer, vs. residing on someone else’s server.
https://www.completeretirementplanner.com/
MaxiFi is my favorite software among those I have tested. Why? Works for all ages, v. strong economic background of 30 year leader (Prof Kotlikoff, Boston U), excellent support, SS benefit optimization, easy to use, inexpensive ($109/yr on renewal).
me, too. Full name is maxifiplanner. The default settings are conservative in the direction of not running out of money (e.g., the default assumption is that one lives to age 99) but all defaults can be varied by the user to explore alternative scenarios or the effect of making various moves. Takes into account all the complicated interactions between the various variables, which is something individuals cannot do with spreadsheets. I highly recommend.
I found FI Calc better and more user friendly than FireCalc. Also use Fido’s planner along with Flexible Retirement Planner.
FIRECalc is what I use. I feel like monte carlo simulations can be useful to show a wider range of mathematical possibilities, but with FIRECalc you are getting results based on how your unique portfolio and spending would have fared during any number of historical periods.
You can plug in current amount saved, retirement length, spending, and other customizations in portfolio composition, expected SS, one-time additions or withdrawals, etc. As an example, lets say you project a 30-year retirement (the range is customizable). The software will model how your portfolio would have fared during 123 different historical 30-year periods going back to 1871. So how it fared from 1871-1901, 1872-1902, all the way up to 1992-2022 (2023 data is coming soon). It will tell you what percentage of the timeframes your plan would have failed.
If you know your portfolio would have survived every conceivable period since 1871, wouldn’t you feel somewhat confident that your plan could survive your own retirement? An analogy on the calculator front page reads:
“Suppose you are building a house in Honolulu. No one could predict the temperature for any given future date during the decades the house will be used. But if you know that it has never been under 52° in that location in all of recorded history, you could make an intelligent judgment about how much heating capacity is enough.”
Instead of a fixed 8% return I’d want to use a ‘monte carlo’ simulation of using historical returns over a retirement period.
For me, personally, my portfolio’s CAGR over the last 30 years has been just a bit over 7%.
That is excluding any contributions made during a given year. So the actual return is probably a bit lower.
Dick, thanks for a thought-provoking article. Every engineer / computer scientist learns early in their career the acronym GIGO – garbage in / garbage out. Faulty information and/or assumptions will lead to faulty results. It doesn’t matter if you are using a software tool, or pen, paper and calculator. All results need to be sanity checked. That’s standard good analytical practice. I use tools as a guide and a check for margin of safety. No, you can’t predict the future. But if you run a series of simulations with conservative assumptions, and your plan shows positive margin of safety, you likely have a reasonably good plan.
If you have ample safe income these calculations lose some importance. But for many retirees, expecting to live on SS and modest savings, a projection might help them make some better decisions. And projections aren’t one and done. I revisit them yearly, and good financial planners do a yearly review/assessment for their clients. Like the weather, it is hard to predict the future. But the closer we get to an event the better we can predict things.
While an 8% annual return sounds fine to build into a forecasting spreadsheet, it’s the sequence of returns risk that can kill a portfolio. Personally I wouldn’t be too giddy about retiring today and starting the withdrawal and growth rates Mr. Quinn mentions with the markets at historic highs. But I’ve always been a bit conservative.
And then there’s the real catastrophic events: nuclear war breaks out, or the sea rises 10 feet, or revolutionaries overthrow the government. You won’t find events like that your software! The guys in Germany who retired in 1938, how did it work out for them?
These real but very, very rare giant catastrophes, like Yellowstone erupting, cannot be planned for. The common term for this kind of event is the Black Swan. Sure, an asteroid wiped out the dinosaurs. The sun could emit a giant flare. Even survivalist with years of dried food would find that their preparations are insufficient. You have to ignore these possibilities in your planning and just plan for more normal events.
While I largely agree, sea rises can be hedged against by property location and other means, individual governments being overthrown can be hedged against through international diversification and other means (I know someone that keeps a small amount of gold and jewels in case he needs to bribe border guards some day), the sun emitting flares could knock out satellites (including GPS satellites)which is why I keep updated paper maps of my state and surrounding states, etc. While not foolproof, going through the thought process of “what if” might leave you better prepared for an emergency than others around you.
Any software projection is better than none, but you can’t think of each of their final projections as anything more than one possible future outcome. When doing a multi-year spreadsheet, to obtain an overall probability for the result at the final outcome, you must multiply the individual probabilities for EVERY single assumption. Even if you assume that every individual assumption is 99% accurate, when you get enough assumptions over a long term future, the math will result in a zero probability that the final outcome will be right.
This is why good planners will take sets of assumptions and run them through Monte Carlo simulations in which they can be tested against many years of historical results and give you a better feel for whether or not your spending and investment results are sustainable into the future.
Every year many very intelligent economists use their own algorithm driven spreadsheets to forecast economic growth for just a single year into the future. Very few are ever correct. Trying to forecast the future is a humbling task.
Thanks, Richard, for these cautions.
I rely solely on my own calculations, and neglect the dire warning that greets me upon each login to my Fidelity account: “Income may last until 2027: You may run out of money before your plan ends.” At my current rate of withdrawal, my “plan” should last well into my 90’s, though I’m only in my 60’s! It seems marketing can work to scare investors, too.
Dick, I’m heartened to see you recommend planning based on estimated expenses rather than salary replacement 🙂
While I agree there’s no need to get too far into the weeds, I’d say one should go a little deeper than you seem to suggest. Beyond considering a mortgage scheduled to end, one might add a number for increased health care costs, or more spent on fun (travel, etc) when time is available to have it, or for charitable or family giving, which one might want to increase. If one is moving, consider cost of living in the new place. Again not way deep into the weeds, but a few things that might have a big impact.
Good point on thinking about whether a retirement portfolio would still be positioned to return what one’s working portfolio might.
Sorry, Micheal, I still feel it is easier and safer to plan on replacing 100% of base working income as the primary goal even as I recognize my decidedly minority view- I actually found three people who agreed with me though.
During our working lives what we have to spend is determined by our income. Why should it be the opposite in retirement? I think it much easier to project income regardless of source rather than future expenses.
If you wanted a fun travel trip while working, where did the money come from, from income you accumulated, right? That’s exactly what we do in retirement. I am in Florida for a month. The total cost is around $15,000. We accumulated that amount in a travel savings account over the last year, just as we did while working.
Anyone can project their SS and/or pension income, estimate dividend or interest income, even estimate income from investments and my strong suggestion, from an immediate annuity. I feel that from whatever source a steady income stream for regular expenses is key.
But again, no matter what your expenses they must be paid from income so that comes first in my book. I doubt many retirees will plan or experience ongoing expenses more than 100% of working income. Even if my goal is overstated you have an inflation cushion.