In all my years of working, I have yet to run across someone who didn’t appreciate getting a raise or become really agitated with the prospect of taking a salary cut. Justified or not, the way that income level sets personal expectations seems to be ingrained in each of us from a young working age. And after thinking that way for perhaps decades, it should come as no surprise that such a mindset doesn’t necessarily immediately evaporate the day we retire. If you have a choice, do you really like the idea of leaving retirement income on the table?
So when studying the ins and outs of retirement finance, one little detail tends to really nag at the minds of certain optimization-oriented people — the assumption about the retirement spending method. You see, the vast majority of retirement research (including the Withdrawal Rates calculator on this site) uses a rather simplistic model for retirement spending that assumes purchasing power remains constant from the day you retire to the day you die regardless of how the markets perform. The goal of these studies is simply to protect you from the worst-case scenarios, while the prospect of making millions in the markets you’re not allowed to touch is not addressed.
While I actually really like the straightforward traditional approach as a sanity check for people with a good feel for the amount of income required for their own happiness, I do appreciate the perspective that there’s more to the retirement story than not running out of money with a flat spending trajectory. In lots of cases, it really is safe to give yourself a raise in retirement. And in others, cutting back a little when the markets do poorly can work wonders for the long-term safety of your portfolio.
There’s actually been a ton of research in this field, with lots of scholarly-sounding algorithms claiming to be the best method to calculate retirement spending. But like lots of investing topics, there are very few good resources for modeling these methods in a dynamic and repeatable way so that you can identify the best one for you and your personal asset allocation. So what does an engineer do when he finds an open calculation sandbox like that? He builds a spreadsheet!
The Retirement Spending calculator is designed to compare a variety of different popular withdrawal methods for any asset allocation you choose. And like every other tool on this site it approaches the problem from a start-date-independent perspective, overlaying the results for the same plan starting in every year since 1972 (the most data we have available for so many assets). The end result is an easily-adjustable snapshot of the historical big picture containing not only the best times but also the worst.
You can see in the above screencap that the calculator has two separate charts. The first maps the inflation-adjusted spending levels for each of our hypothetical historical investors, and the second maps the full range of account balances for all of these investors over the years. Use the first to gauge how the allowed spending varied over time, and the second to keep an eye on the portfolio trends in the background.
The inputs include the asset allocation, starting portfolio value, withdrawal method, and withdrawal rate. Note that unlike the Withdrawal Rates calculator that determines withdrawal rates automatically, the Retirement Spending calculator allows for manual tinkering. I’ve found that you can learn a lot from playing with the numbers and watching how the charts move — give it a try, and you’ll see what I mean. For some spending methods certain inputs may appear and disappear, but experiment long enough and it will all make sense.
The Minimum Required Spending and Minimum Tolerated Balance inputs are completely optional (you can set them to zero to get them out of the way), but I definitely recommend using them. They control the dashed horizontal lines on the two charts, and help identify some really critical data points in evaluating any retirement plan. After all, despite the traditional definition of retirement failure as “running out of money” there are actually three ways that a withdrawal plan can fail you:
- You run out of money
- Your portfolio dips low enough in a volatile market to spook you out of your plan and you abandon it altogether, likely selling low in the process
- Your allowed spending falls below your minimum needs (food, rent, etc) and there’s no more room to cut
Failure #1 can be easily seen in the account balances chart by the blue field hitting the zero line. The SWR in the example above is selected to model the traditional definition of “success” meaning that in the single worst retirement scenario you ran out of money at the end of 30 years.
Failure #2 is marked by the black dashed line on the account balances chart, and is set in this example to a 50% drop in the original portfolio value. Are you the type of person who might get extremely nervous with a 50% drop in real portfolio value just 6 years into retirement? I sure am. If so, maybe a different asset allocation or spending method might solve the problem.
Failure #3 is marked by the black dashed line on the spending level chart. Note that there are no failures in this example because the Baseline % method is specifically designed to never lower spending, but other methods like Constant % and VPW are not so certain. If there’s a real chance of your spending method not supporting your minimum needs, perhaps a new plan might be necessary.
There are a bevy of different withdrawal methods out there to choose from, but for now I’ve started with four — Constant $, Constant %, VPW, and Baseline %. Rather than detailing every option here, I highly recommend reading the Methodology for details and working examples for each. As an example of how the withdrawal method affects retirement performance, here are the charts for each available method for the same Classic 60-40 portfolio and spending/balance requirements:
That’s quite the difference in both spending levels and account balances! I’ve found from my own experiments that certain withdrawal methods work better for some portfolios than for others, and I hope you find the tool equally helpful in exploring all of the possibilities for your own portfolio. IMHO, there’s no such thing as a single superior withdrawal method for every investor and asset allocation.
Just as I aim to expand the portfolio options as I find new ones, the plan is to also expand retirement spending methodologies. If there’s one you would particularly enjoy being added to the list, please let me know. And while I can’t go into every method in detail here (that would be a really long post), I imagine that I’ll write quite a bit about the pluses and minuses of different methods in the future.
Gathering, processing, and sharing information like this is a growing process. There are many great tools for exploring the accumulation side of the equation, and the Retirement Spending calculator will hopefully empower us all to become better and more knowledgeable investors once it comes time to put our hard-earned investments to good use. No matter where you are in that journey, the best time to start planning for the future is today.
UPDATE
The retirement spending calculator has received a large update since this was first written. Check it out!