I have been having a few thoughts lately about withdrawal rates, spurred by a comment a few months ago on my previous post on the subject. Many investors don't realize what usually happens in a withdrawal scenario. When they look at a Safe Withdrawal Rate (SWR) table like that from the Trinity Study, (reproduced below) they simply try to figure out how to have a 100% success rate (i.e. never run out of money.)
With our current low-interest rate environment, and many gurus projecting low future stock returns, some investors even argue that the Safe Withdrawal Rate should be 3.5%, 3%, or even 2.5%. I think that's kind of silly for several reasons I've discussed before and will discuss again in this post.
However, what people DON'T really think about, is just how much money they are likely to have left if they go for a “100% success SWR.” In order to look at that, it is worthwhile looking at a paper by Phillip Cooley in the Journal of Financial Planning.
In this paper, Cooley et al included this table, which shows how much money you have left, on average (technically median), using any given withdrawal rate adjusted for inflation.
This table assumes a starting portfolio value of $1,000. So if you use a 4% withdrawal rate (adjusted each year for inflation) and a 50/50 portfolio, on average, after any historical 30 year period, you would be left with $2,971 at death. That's right, on average, you die with three times as much money as you had upon retiring. Even with a 5% SWR you will on average end up with more than you started with. Surprised? Don't be. That's just the nature of averages. That's why it's called a “Safe” withdrawal rate. It's very safe. You can take out more money, and may even do okay, but that would be a risky withdrawal rate. You know, 6 or 7%.
Six Other Reasons To Avoid Hyper-Conservatism
There are other good reasons not to use some ridiculously low withdrawal rate (like 2.5%.) You all know I like lists, so here's another one.
# 1 You probably won't live for 30 years after retirement.
Remember that if the odds of you running out of money in 30 years are 10%, and the odds of you actually living 30 more years are 10%, then your actual risk is 1%. That's lower than the risk of an experienced mountaineer dying on Mt. Everest (1 in 64, it's 1 out of 5 on K2.) At any rate, if you retire at 65, your life expectancy as a man is 18 years (20 as a woman.) Chances are not great you will make it 30 years. Now if you're retiring at 45, then I think it's a good idea to be a little more conservative, but you get my point.
# 2 Nobody actually follows a fixed SWR
Everybody I know in real life retirement keeps track of how they're doing. If their returns are poor, they cut back, tighten their belt, cancel a few vacations, give less to charity, give less to heirs and make it work. When times are flush, they spend a little more. Retirement withdrawal rate researchers like Wade Pfau are starting to acknowledge this fact in their work, but real retirees like Taylor Larimore have known this for years. In a recent forum post, Taylor explained his simple strategy:
I retired in June of 1982 at the age of 57. We had about a $1 million dollar portfolio to last us the rest of our lives. I didn't know about safe withdrawal rates (the Trinity Study wasn't published until 1998). We had no computers, Internet, Monte Carlo, or sophisticated calculators. We only knew that we had to be careful to make our money last ($1M at 4% = $40,000/year before tax).
So what happened? We simply withdrew what we needed and kept an eye on our portfolio balance. Most years our balance went up and we spent the money on vacations, luxuries and charity. When our balance went down we tightened our belt and economized. This is what most people do and it works.
# 3 Most Retirees Spend Less As They Go
Retirement spending follows a curve where it is highest the first few years of retirement when lots of purchases are made and traveling is done and then gradually decreases until just before death, when it ramps up dramatically with medical and/or long-term care costs. So if 4% (indexed to inflation) provides plenty of money for those first 5 years, it'll probably be more than enough for the 20 after that!
# 4 75% Certainty Is Good Enough
The Trinity Study authors themselves suggest 75% certainty is “good enough.” Even William Bernstein, who is quite conservative with regards to future expected returns, has written that going for anything more than 80% is foolish because there's a 20% chance of your country imploding at some point during a 30-year retirement. Well, 75-80% certainty over a 25-30 year retirement for a portfolio of 50-75% stocks corresponds to an SWR in the 5-6% range. That's 25-50% more money to spend each year in retirement. That's hardly insignificant.
# 5 Your AUM Advisor Has A Serious Conflict Of Interest
Your advisor, especially if paid a percentage of assets under management (AUM), may suggest you only withdraw 2.5-4% of your portfolio each year. However, always remember his bias. First, the larger your portfolio gets the more he gets paid. The biggest threat to your portfolio as a retiree is you spending it! So if he can keep you from spending it, his paycheck gets larger. AUM fees are the best kind of passive income! Second, he's only going to get in trouble if you run out of money and live a long life. If you die with lots of money, nobody is going to get mad at him. But he could really care less how often you go on vacation, how much you give to charity, or how comfortable your retirement is. Higher portfolio withdrawal rates are all downside to him, whereas they are a mix of upside comfort and downside risk with you.
My dad out turning money into happiness flying to his favorite hunting spot
# 6 You're Probably Too Conservative Already
Those who acquire large nest eggs and those who read financial blogs like this one are honestly very unlikely to ever run out of money. It is really hard to go from saving and investing during the accumulation stage to actually spending your money in retirement. My parents have been fully retired for several years now, living on a pension. This year my dad starts taking Social Security. If it wasn't for the IRS mandated RMDs I don't think I'll ever get them to start spending anywhere near 4% of their stash each year. They didn't become affluent by being spendthrifts, and who is going to change habits after half a century of living?
One Invalid Reason
Some think it reasonable to increase the SWR because the original Trinity Study used large cap equities only. However, I think that's probably not valid. Yes, they used only large cap equities, and small cap and value equities probably have a higher expected return. However, they also used long-term corporate bonds- not exactly the short term treasuries that many investors are using today in their portfolios. While none of us really know what to expect out of equities in the future, bond returns over the next 5-30 years are very likely to be lower than they have been over the last 30 given the interest rates we are starting from. Those two factors probably cancel each other out…at best.
The other factor that far too few investors take into account is the effect of advisory and other portfolio fees. If you're paying 1% in fees, your 4% is really 3%. [Update: I was challenged in the comments section below by someone who cited Kitce's study about how a 1% AUM doesn't lower your SWR by a full 1%. I found the argument weak in that it certainly DOES lower your withdrawal rate by 1% of your assets in year one and that whether it lowers the SWR by 1% or not, it still lowers the amount of money you have to spend by the amount of the fee, which is really the point of my statement anyway.] Why any retiree would consent to have 1%+ AUM fees assessed against their portfolio when there are asset managers willing to do it for far less (or even a flat annual fee of just a few thousand) is beyond me, but I'm sure there are plenty of retired docs out there paying $20K+ a year in asset management fees.
The Bottom Line on Safe Withdrawal Rates
As I've said many times, the point of the Trinity study wasn't that the SWR is 4%, it was that the SWR is not 8%, 10%, or even 12%. Start with something around 4% and adjust as you go. Staying flexible is immensely valuable both in increasing your ability to spend as well as decreasing your likelihood of running out of money if you are lucky (unlucky?) enough to live into your 90s. Better yet, put a floor under your spending with Social Security, pensions, and perhaps even a SPIA or two and then have at it with your portfolio withdrawals.
What do you think? Were you aware that most who followed a 4% SWR rule die with more than they started retirement with? How are you (or how do you plan to) managing your retirement withdrawals? Comment below!
While I agree with you that advisory fees are an important consideration, I think you are incorrect to state 1% aum fees will reduce your swr from 4% to 3%
https://www.kitces.com/blog/the-impact-of-investment-costs-on-safe-withdrawal-rates/
Interesting perspective. A lot of handwaving there that basically says “but the portfolio (not the client) is paying that expense” and the portfolio can afford it because most of the time you’ll die with tons more than you started with. The fact remains that there is 1% of the portfolio that you could spend on something else if you weren’t paying an advisor.
What Kitces is saying there is that if you’re willing to be your own advisor is you can take 4% of the original retirement portfolio adjusted each year to inflation or you can take 3.6% of the original retirement portfolio adjusted each year to inflation PLUS 1% of the current portfolio if you’re willing to do it yourself or pay that 1% to an investment advisor. Three options there but the way he phrases it makes advisors come out smelling like a rose.
Hand waving or not, he is correct. 1% in fees will cost you 1% of your assets under management per year (as you accurately state in the above comments) , but it will not reduce your SWR by 1% (as Kitces states). It is not a perspective…it is math. Two totally different measurements. Both true. Your article is misleading with respect to this extremely important point and I think it is critical that your readers understand this. The difference between 3% and 3.6% SWR is huge! Your articles are a fantastic service to the physician community and I am grateful for the work you do, but they should be mathematically accurate.
PS: I am not a fan of high fees, but some would be best served paying 1% for an excellent adviser. If one can find someone excellent who provides the same level of service for lower than this then all the better. Many docs I know have so little financial acumen it is scary (they should all read your site in my opinion). The real value of an adviser is not in the asset management side of things, but in the behavioral/life coaching aspect and protecting from financial predators.
The problem is once you can hire an advisor and know HE isn’t a financial predator, you don’t need help telling who the predators are.
As far as the “SWR concept” I find the argument a bit irrelevant as the 4% rule is a ballpark rule anyway-i.e. to get you in the right ballpark. If you’re wise, you’ll start there and adjust as you go. If things are going well, you can spend more. If poorly, then you can spend less. But I think this “mathematical proof” that says “a 1% AUM advisor only reduces your SWR by 0.4% (or whatever)” functions more as a justification for what is in reality an excessive fee for asset management than as a serious method of spending your retirement money. 1% of a physician’s retirement each year for asset management is simply too much money and there is no justifying it, whether it reduces your SWR by 0.4% or 1%.
Well I think we are talking past one another here but I stand behind my reasoning. I agree with many of your points but disagree that SWR concept is irrelevant. If one decides to retire when they reach a 3.6 SWR vs a 3.0 SWR this is the difference of years or additional working for most people with less than optimal savings rates. It’s the difference between thinking one has to save 3m vs 3.6m! I know SWR is not perfect, but I can’t think of a better starting point to decide when enough is enough. If I was paying a 1% AUM and I got to around a 3.6 SWR I would consider work optional at that point. There would be no mathematical need to go to 3.0%
But the 1% is really irrelevant. I think it is distracting from the real issue. Lets call it 0.5%AUM. Is that an acceptable rate? It doesn’t reduce your theoretical SWR from 4% to 3.5%. Whatever you feel is an appropriate amount of a AUM to pay for financial advice, your statement in your post is not mathematically factual.
By the way, 1% AUM for asset management only is absurd. If that is the only thing one is using an adviser for that’s way too much. If one needs only asset management get a roboadvisor. At 1% I would expect someone to work with me all through residency when they were making almost nothing off me and be a full service adviser/financial coach. I can think of quite a few of my colleagues that would have come out ahead had they been paying a good adviser 2 or 3% AUM (Not that I would ever recommend this insane amount for anyone). I know more than one person who got out at the bottom and never got back in due to fear. They needed a trusted adviser at any price.
BTW: I’m sorry for belaboring this point, but please take it as a compliment. If I didn’t care about your content or your message I wouldn’t be wasting my time commenting. When you use detailed mathematics to argue for bonds being in a taxable account (which is an excellent article)for example, or to rip apart crappy insurance products you set the bar high for me to expect mathematical accuracy in other posts.
I agree it is important for deciding when enough is enough.
I updated the post a few days ago to reflect your points, including a link to Kitce’s article.
The 1% reduction IS factual the first year of retirement, although as you noted abnormal market returns (i.e. when the SWR concept matters) can decrease your fees over time.
But the more important fact in my view is that whatever you choose to pay an adviser is money you cannot spend on yourself, your heirs, or your favorite charities.
I think the decision to focus on SWR or the AUM% may come down to what a person values. If avoiding running out of money is primary concern then knowing your SWR is very important. If you have no worry about this and want to maximize wealth then it makes sense to focus on total return (which as we know is highly influenced by fees). For me it is of high value to know if my SWR is 3%, 3.5% or 4% but I can imagine for some it may not really matter. The great news is that lowering fees increases SWR so I 100% agree with you that minimizing fees are critical! Fees are money that cant be spent on cool stuff like rock climbing 😉
One point that may be very important is expected rate of return. If one has a more conservative portfolio or is expecting lower than historical return going forward then the AUM% has a greater negative impact on the SWR (I think). Don’t know if I’m smart enough to quantify this but it would be an interesting analysis. A heavy bond or fixed income portfolio would be a disaster with high fees.
Thanks for the discussion and the update.
One more point for your readers (if anyone is still reading this, haha): You have to add transaction and expense ratio to your AUM. If your average mutual fund cost is 1% and you are paying a 1% AUM these both need to be factored in. I think all the SWR research assumes no transaction or annual expenses which is unfortunate and a but misleading unless someone really digs into the studies.
PS: Throw in more cool pictures you take while rock climbing…it is inspiring!
I think you’re putting too much precision on SWR that isn’t really there in real life.
I don’t think the ERs matter much if we’re talking about 5 or 10 basis points.
Correct, 5 or 10 basis points don’t matter. Not everyone reads your blog though and many pay 10X that ER.
I don’t think I’m putting too much precision on SWR…I think more likely I have failed to write clearly. Arguing through blog comments is not the greatest format for debating a nuanced topic. I’m happy to continue this discussion if you wish. You have my email.