This is Part 2 of Required Minimum Distribution Follies. See Part 1 here.
Oh, the Safe Withdrawal Rate dilemma. There’s been a lot of words written about this. Here are some more.
What is the point of the Safe Withdrawal Rate (SWR)?
The whole idea is to figure out how much a retiree can take out of their accounts annually without later running out of money. This of course assumes that the the retiree can live on the amount of withdrawn money. If the retiree needs more than the SWR, then there is a potential future problem of “insufficient funds” (namely, running out of money).
This post is an investigation into whether the Required Minimum Distribution (RMD) can be used as a Safe Withdrawal Rate.
While a retiree can possibly adjust their standard of living to meet some theoretical percentage goal, there are things that cannot be adjusted, and most are outside of a retiree’s control. One primary immutable factor that needs to be considered is the actual amount of their RMD. This value is set by IRS regulations. As a result, the RMD could possibly be higher than a retiree’s ideal withdrawal rate. Or lower.
We have no direct choice on how what our RMD is. When we hit 70-1/2, RMD is determined by IRS rules, and is calculated based on the total amount in your tax deferred accounts, and your age. The first year withdrawal rate is 1/27.4 of the balance in all your tax-deferred accounts which is approximately 3.65% (actually 3.64963504%; all other percentages and numbers mentioned here are rounded to two decimal places). For more details on when the RMD must be calculated, see Part 1 of Required Minimum Distribution Follies here.
(I supposed I could say that we do have some indirect choices. We could have chosen to put less into tax deferred accounts, or convert some tax deferred amount to Roth IRAs for example. And while I will be converting some tax deferred amounts to Roth over time, to avoid a higher tax bill, I will do this slowly over years, and it will not have an initial significant influence on my overall situation. But this is so indirect, so unlikely to affect my tax deferred account balance in any meaningful way in the short run.)
Most of the debate around the SWR involves what is an appropriate percentage of money to withdraw to avoid the dreaded running-out-of-money boogeyman. Four percent is most often said to be the correct amount, with some saying that dropping that down to 3% is more conservative, namely even safer.
Since the IRS requires everyone to withdraw at least 3.65% as RMD from their tax-deferred account their first RMD year (ie, calculated from the December 31 prior to turning age 70-1/2), then that could be seen as a potential SWR. That is assuming that the amount in tax-deferred accounts is large enough such that 3.65% of it is enough to live on.
As it happens, as I approach my first RMD, I used some of the free online calculators to approximate how much RMD I would need to withdraw to satify the IRS rules. It turns out that for some people, the first year’s RMD amount will be higher than had been expected. It happened to me.
But the extremely surprising fact is that the RMD percentage increases over time! The actual IRS RMD calculations use a slowly increasing percentage. Not only do the RMD percentages increase each year, but the rate of increase increases as well. The percentages are (supposedly) based on life expectancy. The older you are, the less your expected lifespan, so (for some reason) the IRS wants you to reduce your account balance.
The following graph, reproduced from Part 1 here, shows the RMD percentage for each age. At the age of first RMD withdrawal, the percentage is about 3.65%. By age 79, the percentage is over 5%. By age 93, the RMD percentage is over 10%. If you are “lucky” enough to live until age 115, the RMD percentage is over 50%.
If the balance in the tax deferred account is static, namely if the assets in the account does not appreciate, then because the RMD percentage increases, the remaining balance obviously decreases. The decrease in assets is enough to cause the actual RMD amount to decrease over time even though the RMD percentage increases.
If the assets in the retiree’s tax deferred accounts increase (let’s say you hold mutual funds and the stock market goes up), then the balance of remaining assets can increase for some years, and the RMD amount increases concomitantly. (Oh, how I have wanted to use that word for years, and finally have a chance.) However, in most cases, the remaining balance eventually declines because the RMD amounts being withdrawn are great enough to eventually overtake the increase in underlying asset value.
Scenario 1. Follow the money: standard RMD withdrawal.
As mentioned, the first year RMD percentage is approximately 3.65%. Let’s take a hypothetical situation of say, a $1,000,000.00 balance in a retiree’s tax-deferred accounts, assuming the account is static and assets do not increase. The first year’s RMD amount is $36,496.35, so the balance of the account is reduced to about $963,503.65. The second year’s RMD percentage is 3.77% so the RMD amount (3.77% of $963,503.65) is $36,358.63 which reduces the account balance to about $927,145.02.
As noted the RMD percentage increase, the RMD amount decreases, and of course the account balance decreases. By the age of 80, the RMD percentage has risen to 5.59%, the RMD amount has decreased to about $33,974.25, and the account balance is now at approximately $574,164.81.
So the RMD amount decreased several hundred dollars per year. The account balance is now at about 57% of it original value.
By age 90, the RMD percentage has increased to about 8.77%, the RMD amount is near $28,000, and the account balance is dipped below $300,000.00.
If you live to 100, your RMD percentage is up to 15.87%, the RMD amount is somewhat over $14,000.00 and your account balance is down to somewhere near $79,000.00.
Anyone living to 115? The RMD percentage is a whopping 52.6%, and the RMD amount is slightly more than $300.00 and the account balance is a similar amount.
The ongoing decline in RMD amount reduces the retiree’s standard of living as they get older, and the ever-declining account balance is the opposite of comforting.
The following chart shows the values for this scenario.
Columns: age, IRS value, RMD percentage, and RMD amount for a $1,000,00.00 account, and the balance remaining in the account after the RMD withdrawal for that year.
Scenario 2. Suppose we use the first year’s RMD percentage as a theoretical continuing SWR.
A static 3.65% is not realistic nor is it a legitimate withdrawal rate because IRS mandated RMD percentage increases each year. But let’s look at its theoretical implications. If only in theory, let’s see how that would turn out.
If one takes the first year’s RMD percentage and uses that 3.65% as a constant withdrawal rate, then one must be aware of the fact that 3.65% of a declining asset is an ever-decreasing number. Let’s start with the same $1,000.000.00 static balance, where assets do not increase. The first year’s withdrawal is about $36,496.35, leaving a balance of $963,503.65. The second year’s 3.65% withdrawal would be $35,167.75 leaving a balance of $928,335.77. While the difference between $36,496.35 and $35,167.75 is only $1,190.75, it indicates that each year the retiree would have a smaller and smaller withdrawal amount than the year before, and their base account balance would also be decreasing.
By age 80, the approximate account balance is $608,000.00, and the retiree’s annual 3.65% percent would amount to only somewhere near $25,000.00. That amount is a more than $11,000.00 decrease from the same percentage at the first withdrawal. By age 90, the retiree’s 3.65% withdrawal is something near $17,000.00, less than half their original withdrawal amount. And the account balance is somewhere near approximately $293,000.00.
The following chart shows the values for this scenario.
Columns: Age, IRS value, correct RMD percentage, the withdrawal amount of 3.65% it it had been legal, the remaining balance. “Diff 365-rmd” is the difference between the static 3.65% and the actual correct RMD amount. Data through age 100 is shown.
* * *
Both options 1 and 2 all point to the fact that it is likely that the retiree would either need to draw on other assets, or reduce their standard of living so dramatically as to be eventually just about untenable.
Scenario 3. Let’s see what happens if the retiree withdraws the same amount every year.
What would happen if one takes the first year’s RMD amount and withdraws the same amount every year (as opposed to a static percentage or an increasing percentage), starting with the same $1,000,000.00 balance. This option is more real-world because the IRS RMD amount declines each year (even though the percentage increases), and in this case after the first year we are withdrawing more than the RMD amount.
The first year’s withdrawal is $36,496.35, leaving a balance of $963,503.65. Withdrawing the same $36,496.35 each year (assuming the account’s assets do not increase) means the retiree will run out of money at age 97. This scenario might provide a much better standard of living than the previous two examples, assuming that the retiree has other assets, or dies on or before age 97.
The following chart shows the balance in the account after RMD withdrawal of the fixed amount each year. At age 97, it’s all gone.
Columns: age, and “bal” is account balance after the static amount of $36,496.35 is withdrawn.
* * *
All three of the above examples, however, do not account for an account where the assets increase, such as in holding stock or mutual funds, etc.
So what does this all mean?
RMD withdrawal is a legal requirement. (See Part 1 here to see what would happen if you do not take your full RMD for any year; not good things happen. Do not take less than the required amount.)
The initial RMD percentage (3.65%) sounds like a pretty fair SWR. If one can live on that amount, all the better. But over time, the decline in RMD amount means it becomes useless as a guideline to expect to live on.
Neither the RMD nor the SWR can take into account anything about what a retiree might need to live on. RMD and SWR know nothing about taxes, inflation, expenses, medical costs, lifestyle, vacations, roof replacements, or stock market declines. So neither are useful for real world planning financial needs, in my view.
Scenario 4. What if the assets increase in value?
The following example is similar to example 3 where the account starts with a $1,000,00.00 balance and the same $36,496.35 is withdrawn each year, but also assumes a 3.65% increase in the account (such as mutual funds in the account increase in value). In such a case, the account will run out of money when the retiree would theoretically be over 160 years old (!). No chance of running out of money during a normal lifetime or two. A certainly more pleasant outcome.
The following charts show what the balance in the account will be when the $36,496.35 amount is withdrawn each year and assumes there is a 3.65% increase in value of the account each year. As in previous scenarios, the account starts with a $1,000,000.00 balance.
Columns: age, “bal”is the account balance at start of year, “Minus 36496.35” is the balance after the withdrawal amount of $36,496.35, “If 3.65%” shows the 3.65% of the then current account, “Plus 3.65%” shows the balance when the increase of 3.65% is added to the balance. So the “Plus 3.65%” value of one year is the same as the balance starting the next year (“bal”).
It should be logically obvious that a steady 3.65% increase in value is not realistic. Stock markets do sometimes decline and sometimes decline for extended periods, not all companies increase in value the same amount, some companies in which you have invested may decline or bankrupt, real estate does sometimes decline, natural and unnatural disasters happen, assets do not always appreciate on schedule. The long term stock market gain (S&P 500) is something close to 10%. This is of course more than twice as large then the 3.65%. But since we are required to withdraw RMD even if the market is down, we cannot rosily plan for a smooth increase in asset values.
Is there a real real-world scenario available?
Perhaps there is one. I would surmise that including two other variables would make these scenarios more realistic and usable by humans. One is to include an adjustment for the actual inflation of any one year, since the value of money in an inflationary world decreases its buying power but likely increases the remaining account balance; in such a case the amount to be withdrawn is increased to compensate for the change in the decline of the value of money. Second is to not expect a static annual increase in asset value, but to adjust each year based on its real world change in value in that same year. This might indicate what a real world withdrawal should be. For better or for worse, neither of these two variables can be known in advance.
Almost The End.
Conflating RMD with SWR sounded like a good idea at the beginning, but if you’ve read this far, I would guess you would agree they have nothing to do with each other. Or should not, anyway.
This whole exercise makes me consider if it might have been better to save all/most/some/more money in after-tax accounts rather then tax-deferred accounts. Money in after-tax accounts are under my own control. RMD withdrawal requirements are not. (Happily, and maybe inadvertently, not all of my savings are in tax deferred accounts.) On the other hand, I am under no obligation to spend all the money I withdraw for RMD purposes. On the third hand, I can withdraw more money than the RMD number; after all it is a Required Minimum Distribution, and of course I am under no obligation to spend all of that money withdrawn. Or, if I have other assets, I can rely on them to make up the difference between the RMD amount and what is needed to live on. On the fourth hand, since the RMD amounts decline greatly over time, it becomes less and less likely I would have any withdrawn-but-unspent RMD funds.
If you recall, all the discussions about whether one should put money in tax deferred accounts or after tax accounts, like Roth, centered around whether one thought one’s tax bill would be lower after retirement than before.
Well, I for one, thought that it naturally would be lower. In fact most articles about this topic mentioned that most people felt the same. But now I wonder if that was as wise as I had considered.
I’m beginning to think that the safe withdrawal rate decision is a serious consideration for academic discussion. Since I am not living in an academic discussion, I am in a different place, it’s called the real world.
This is where Social Security can be really useful in making up the shortfall between RMD and what is needed to live on, assuming Social Security still exists if the plunderers in Congress ever stop with their never-ending wrath of destruction. While Social Security increases periodically due to an inflation adjustment, it likely will never increase as much as the RMD amount decreases. RMD percentages are not linked to inflation. And the measure of inflation used to calculate Social Security increases is under political attack in an attempt to make it easier to keep benefit increases smaller or nonexistent.
So, if in the early years of one’s RMDs, the amount of RMD plus Social Security is greater than the amount to live on, it is probably best to save that difference for the time when one’s RMD decreases substantially.
(Please report any math errors you might find.)
The post Are You Ready? Required Minimum Distribution Follies (Part 2) and The Safe Withdrawal Rate appeared first on Smile If You Dare