From time to time the big mutual fund/brokerage houses come out with a study using the data they obtain from their 401(k), IRA, and/or brokerage data. The conclusion of the study, unsurprisingly, is usually that people need to save and invest more, preferably with their firm. While that is probably true, there always seems to be a bit of a conflict of interest there.
Fidelity recently did a similar study, but they just looked at their physician clients. The conclusion, as expected, is that doctors need to save more and get more financial advice. But the really interesting stuff in the study comes when you dig a little deeper. I applaud the Fidelity analysts and authors for doing this study. After speaking with David Martin and MeeJin Annan-Brady of Fidelity recently, I think they probably did about the best they could within the limitations of available data. I disagree with a few of their methods (and conclusions), but overall I think this study makes a valuable contribution to what is known about physician personal finance and investing habits and I applaud their efforts.
Fidelity, as one of the largest mutual fund companies, 401(k) and IRA provider, and brokerage, has access to a lot of physician-specific financial data. They included data from over 5,000 physicians in this study, although a significant downside of their data set is that almost all of it comes from large non-profit health systems, both academic and non-academic. The really nice thing about their dataset is that for certain portions of the study, they only looked at doctors who had been with that particular employer for a significant amount of time, and included not only 401(k) data, but also IRA and brokerage data. For the first time, we've actually got a decent look at what physicians are doing with their retirement savings. In some ways, the data is concerning, but in others, I find it quite reassuring!
Savings Rates of Physicians
The first figure in the study, which I found quite reassuring, was the calculated savings rates for these physicians. On average, docs save 13.1% of their income in their 30s, 14.3% in their 40s, 16.3% in their 50s and early 60s, and 18.3% in their late 60s. On average, the figure is 14.9%. It isn't quite the 20% I recommend, but it's a heck of a lot higher than the 5-10% I expected it to be, and I found that reassuring.
Maxing Out The 401(k)
The second figure in the study shows that only 40% of doctors under 50 hit the “402(g) limit” ($17.5K into a 401(k) or 403(b).) That number rises to 70% for doctors over 50. Personally, I think that's terrible. The average doctor in this study makes $299,000 per year. $17.5K is less than 6%. If they're saving 13.1%, where is the other 7% going if not into the 401(k)? For someone over 50, the limit is $23,000, or less than 8%. Where is the other 8% going if not into the 401(k)? Max out your 401(k) people. Seriously. I know it can be kind of tough when you have a $52K limit, but not hitting the $17.5K limit? That's pathetic.
Higher Income Docs More Likely to Use 457(b)s
The third figure shows that doctors who make more than $500K are 12 times as likely to use a non-qualified plan (like a 457(b)) than a doc making less than $150K. I don't think anyone finds that surprising. What blows my mind, however, is that there are 26% of doctors making over $500K who aren't maxing out these non-qualified plans. Again, that's pathetic. Don't complain about your tax bill being high if you're aren't willing to save 7% of your income toward retirement (7%= ($17.5K 401(k) + $17.5K (457))/$500K). Fidelity's conclusion from this data is that more employers should offer 457s. I guess that would be nice, but my reading of the data is that docs either don't want them or don't know about them, for better or for worse.
Doctors Taking On Too Much Risk As They Approach Retirement
The fourth figure in the study is the one getting all the attention. This shows that physician asset allocations (and again, according to David Martin, this data reflects only qualified plans and brokerage accounts, not IRAs due to technical issues with obtaining the data) actually get MORE aggressive the closer they get to 60. On average, only 6% of docs in their 30s have an “aggressive” asset allocation but 42% of doctors in their early 60s have an aggressive asset allocation. This suggests that physicians are trying to “hit a home run” at the end of their accumulation phase, possibly due to lack of adequate savings early on and possibly due to inadequate financial advice/planning. Mr. Martin suggested the effect would be even worse if IRA data were included as their studies show that IRAs on average are more heavily invested in stocks than 401(k)s.
However, it is important to realize what they are really showing here. The “aggressive” category is an asset allocation that has more stocks in it than the Fidelity Freedom fund for their age. The 2020 fund, for a doctor who wants to retire in about 6 years, would be the corresponding fund for a doctor in his early 60s. It has an asset allocation of 61% stock. If you have 71% or more, you're aggressive according to the study. That seems reasonable.
But what about the younger docs. If you're 35, you presumably should have an asset allocation corresponding to the 2045 fund. That fund says you should have a 90% stock portfolio. To beat that by 10% or more means you have a 100% stock portfolio. Is Fidelity really suggesting that more docs in their 30s should be MORE aggressive than that? Likewise, if you only have 80% stock (a very aggressive portfolio according to most experts) you're in the “conservative” category according to Fidelity. Now you can see that the chart really doesn't mean much at all, and should probably be ignored. It would have been a lot more interesting to just see what the asset allocation was for each age group, rather than having it compared to a Freedom fund. It would have been even more interesting to compare the asset allocation for a group of doctors in 2009 vs 2014. Maybe in their next publication. The conclusion from this data is that doctors need more financial guidance. While that may or may not be true, the data certainly doesn't prove that. Comparing average asset allocations to Fidelity's idea of a proper glide path doesn't really give you sufficient data to draw any kind of conclusion in my opinion.
Doctors Can Replace 56% of Their Pre-Retirement Income
The last part of the study was the one I had the biggest problem with. In this section, they took what doctors had in IRAs, brokerage accounts, 401(k)s, defined benefit plans, and Social Security and calculated out that doctors could replace about 56% of their income. They then somehow estimated that doctors needed 71% of their pre-retirement income replaced and concluded that doctors need to save a ton more money, preferably with Fidelity. I saw that 56% figure and said, “Awesome!” My calculations show that most doctors can have a very comfortable retirement on just 25-50% of their pre-retirement gross income. Basically, I subtract out all the stuff I won't have to pay in retirement like retirement contributions (20%), payroll taxes (5%), mortgage (10-20%), disability and life insurance premiums (2%), child-related expenses (5%), work related expenses (2%), college savings (3%), most of my income taxes (15%) etc. I would have a dramatic increase in spending if my retirement savings replaced 71% of my current income. Now keep in mind the 71% figure Mr. Martin uses includes Social Security, and my 25-50% number does NOT include it. But even once you count that in, there's still a pretty big gap.
So I asked Mr. Martin about this. He stood behind his number, and pointed out a couple of reasons why he thought his number was right and mine was wrong (although he admitted, as I do, that every investor needs to run their own estimate.) The first reason was that when you look at the averages from studies on spending that the government does, many retirees are still paying mortgages (or even rent). Of course, they weren't able to actually tease physician-specific data out of these studies, so they generalized that many docs were also still paying mortgages in retirement. (A dumb idea if you want my opinion. I think you ought to have your house paid off within 15 years of finishing residency, long before retirement.) The other reason was that he wasn't calculating a “bare bones” retirement, but more the type of lifestyle people really wanted. Well, his definition of bare bones is certainly different from mine. 50% of a $299,000 salary is about $12,500 a month. 71% is $17,700 a month. I don't know what you're spending every month, but I could donate half of either of those amounts to charity and still do everything I want to do in retirement. I'd love to hear what Mr. Money Mustache would have to say about that kind of spending. Heck, on $17,700 a month I could buy a new car every quarter just with the money I didn't need to buy everything I want and go everywhere I want to go. The truth is that the amount of money you need in retirement has absolutely nothing to do with your pre-retirement salary and everything to do with your pre-retirement (and post-retirement) spending. If you really do need 71% of your pre-retirement salary in retirement, then you'd better do as the study recommends and start saving a lot more.
Lessons From The Ratios
Another interesting statistic comes from looking at Figure 5. The ratios of personal savings (IRA + taxable accounts) to employer provided defined contribution plan to employer provided defined benefit plan to Social Security were very interesting. A large percentage of physicians in the study apparently still have a defined benefit plan, although many cannot actually contribute to them anymore. I'm not sure this data is generalizable to most physicians as it seems very few of us anymore have a defined benefit plan. But the percentages of retirement income from each source were as follows:
| Doctors | Non Doctors | |
| IRA + Taxable | 25% | 15% |
| DC Plans | 38% | 23% |
| DB Plans | 16% | 15% |
| SS | 21% | 48% |
You expect the lower earners to have more of their retirement income provided by Social Security. That's just the progressive nature of the SS program. If you take that out of the picture, you can see that doctors have 48% of their money in their 401(k)s and 403(b)s, 20% in defined benefit plans, and 32% in IRAs and taxable accounts. I find that interesting, but don't know that it actually means anything, and actually have a hard time reconciling it with the data noted above that a huge percentage of doctors aren't even maxing out their 401(k)s.
Overall, I think it is great that Fidelity went to the trouble to look at all this data. However, I think significant flaws in the study design prevent the data from being more useful to doctors and advisors than it could have been. My personal conclusion is that doctors aren't doing nearly as badly as everyone always seems to think they are. Could they stand to save a little more? Of course, who couldn't? Could they benefit from a little more financial planning, either on their own or with a good advisor? Almost surely. Should they be maxing out their 401(k)s from the beginning of their careers? Certainly. Are their asset allocations too aggressive? I don't think we can really tell. But it will be a rare doctor eating Alpo in retirement.
What did you think of this study? Did any of the results surprise you? Comment below!
Another great post. It should be obvious that a Doc should max out all available tax deferred options available to him or her. I personally saved a lot of money in non-tax deferred accounts in my peak earning years. Even if you have to pay some capital gains taxes along the way it is better to save than to spend. Another great thing about saving outside of tax deferred accounts is that there is no limit. I actually now have more money outside of my retirement plan than inside it.
I used to work for EMP (Emergency Medicine Physicians). They use Fidelity exclusively for their 401K. The fund options are extremely weak, Spartan 500 being the only low expense option. I personally think its they take advantage in this regard because the doc are all high income earners and they figure they can extract the most in fees, plus most docs don’t know any better. They definitely have ulterior motives when it comes to stuff like these studies. That being said I agree there are some interesting findings, but don’t be fooled Fidelity isn’t doing anything out of the goodness of their hearts for physicians.
liked the article. really enjoy your site. read your book and enjoyed it. this article on retirement brings up a question. I am a big Bernstein fan, am an er physician, 50, and have nearly “won the game” as william bernstein calls it. are you of a similar theory of taking assets out of risk once the game is won? Or would you just keep 10% in cash for the down years, with the rest in S&P? What is your plan when you win the game?
I am 56 and in OB/GYN. I too am a Bernstein fan. I think I have “won the game” or hit my number. I am interpreting his his take the risk off the table by quitting OB. I am also gradually increasing my Muni bond allocation.
53 years old, anesthesiologist. I also am big fan of Bernstein investment philosophy. I have also hit my “number” as Berstein refers to.
I have personally been paring down my equity positions over the past 24 months as I saw the “number” approaching. I do not want to risk another 2008-2009 financial instrument meltdown and therefore have to continue working, call, weekends etc.
I still have equity positions, but much less than previously. In the words of Jim Cramer, “bulls make money, bears make money and pigs get slaughtered.” I am not going to get greedy.
Although I can’t stop working forever, I feel like I’ve won the game already. I have a job I enjoy, enough insurance that my family’s lifestyle wouldn’t change if I died or was disabled, and have more than enough money to spend. Now the question is what asset allocation I should have. I was pleased with how my 75/25 portfolio performed in 2008, so I anticipate keeping that for quite some time. I expect I’ll probably buy some SPIAs in the 60-70 year old range. But for now, it’s full-steam ahead.
Here’s the problem with the “stop playing when you win the game” plan. Does that mean once you hit enough money that you never have to work again that you just buy a bunch of TIPS and SPIAs? Well, the truth is that amount of money is much higher than the number that will allow me to never work again if I’m willing to continue to take risk in equities. Safe investing is just that- more safety, lower returns. I guess I’ll be surprised if I EVER go less than 50/50. But I agree that it would be foolish to be taking so much risk that you might have to go back to work after you’ve retired. If you’re going to de-risk your portfolio, I think now, after 5 years into a bull market, is a pretty good time to do so.
WCI, can you explain this phrase a little bit: “Max out your 401(k) people. Seriously. I know it can be kind of tough when you have a $52K limit, but not hitting the $17.5K limit? ”
I know what you mean by the $17.5k limit for 401k/403b, but what is the $52k limit?
That’s the limit that can go in including the employer match. Most profit-sharing plans, Individual 401(k)s, and SEP-IRAs have that higher limit.
A few observations:
1) are they including residents as physicians. A lot of the big hospitals that offer Fidelity plans have a lot of residents. This may skew the numbers a bit, and may explain why some physicians are not maxing out their 17,500 limit (I know I should have as a resident, but I didn’t because it was a lot harder to)
2) From my understanding of some 457 plans, if the hospital goes bankrupt your money in the 457 can become available to creditors. I would rather pay the taxes on that money and keep it in a taxable account rather than risk a total loss in the current healthcare environment…
After reading the link I think my first point isn’t really relevant. Still shows people with high incomes and older individuals under contributing. But would still be curious on people’s thoughts on the 457 plans.
I think they did exclude resident physicians. You’ll notice there was no 20-30 year old physician category.
Personally, I wouldn’t worry so much about the 457 risk when I’m staring at a huge tax break. But I would spend that money first in retirement.
SJ, I’m a 53 year old internist and I contributed to a 457 for a year and then stopped when I discovered the fact that you point out: if my non-profit hospital employer goes under, I end up in line with other creditors for that money (or so I’m told). I desperately would love to delay taxes on another 17.5K but just can’t bring myself to risk it given the unknown financial health of my employer (unknown to me).
Thanks WCI and HM. HM I tend to agree, maybe my risk tolerance is very low but I am in my mid 30’s and think the risk, or at least constant worry, of my money disappearing over the next 30+ years outweighs the tax benefit for the 457. I think I’ll still take full advantage of my other tax saving opportunities (403b, backdoor roth, HSA, 529…) but still utilize a taxable account. I’ll just tell myself the advantages of the taxable account (outlined nicely by WCI in another post) make up for the lack of tax benefit. Maybe if I am so fortunate that my taxable account grows nicely and my income goes up more, I’ll reconsider the 457.
How about designing a similar study using Vanguard’s data? Perhaps highlighting those who ascribe to a “boglehead” philosophy vs. active management? That would be interesting! Depending on the data and results, it could be a good marketing tool for Vanguard.
Sounds fun. Do you have access to the data? Perhaps you could lobby Vanguard to do something similar.
I don’t, but we do use Vanguard Asset Management Services for my group’s profit sharing plan. Perhaps I will ask my advisor if they would be interested. I’m sure they could ferret out this data.
WCI…
Have you run the numbers on what $12,500 would be with a 3% inflation 20 years from now(when many of us will be looking at retirement)? By my math $12,500 today would be about $22,000 or with some reverse math the equivalent of about 7,500 today….
I agree with you that retiring on 12,500 would be more than sufficient today especially if you have paid of your debts. However 7,500 looks a lot different. 7,500 is still a lot of money to live on(assuming debts are paid) in retirement today, however many physicians would not be able to live on that amount due to there lifestyle.
The 3 things that concern me the most in these scenarios are 1. Taxes 2. Inflation 3. Age of retirement
Saving as much as possible helps in to combat all of these, but inflation and taxes are killers!
Are my numbers wrong or am I making incorrect assumptions? Can you discuss inflation in more detail in this or a future post?
THanks
There’s no doubt you have to adjust for inflation. $2M today is definitely not the same thing as $2M in 20 years. However, withdrawal studies like the Trinity study generally include an inflation adjustment each year (it’s 4% of the original portfolio plus inflation), but you still have to arrive at the beginning of the withdrawal period with an inflation adjusted number. To make things more difficult, all of our personal inflation rates are different. No one has a personal inflation rate that tracks CPI.
Remember that there are no Payroll taxes in retirement. For self-employed docs this means an automatic 15% savings (yes, I know 1/2 of SE taxes are deductible.)It is unlikely that after contributing the max to Soc. Security and Medicare for 25+ years, you will any ROI for your money. Your Medicare payouts will be just the same as persons who haven’t contributed anything into the System.
I find it near inconceivable that physicians would go from full time work to complete Retirement overnight. Most physicians would find some employment to allow continued IRA/401k/HSA contributions with business deductions even while drawing down Retirement savings.
Look at i-orp calculators to minimize taxes in Retirement.
Inflation is not the boogie-man everyone thinks…Soc Security benefits are indexed to inflation, tax brackets go up. Rarely does everything you want to buy goes up the same. A fixed rate long term mortgage is an excellent inflation hedge.
Early retirees should probably be at the lowest Stock allocation of their investing lifetimes. Wade Pfau recommends a “rising equity glide-path” after that.
Why would you continue to make IRA contributions while withdrawing IRA money? I can’t think of a way that makes sense unless you’re working past 70 and have RMDs. I agree it is often a better scenario to ease off into retirement for many reasons, but not always possible, especially for a practice owner.
Great study, but another major pitfall is that they only had access to their retirement accounts. Individuals may have had retirement accounts at other institutions, or even more likely, it does not account for retirement savings in taxable accounts. The study likely excludes hyper-savers given that they save above and beyond the maximum that can be put into non-taxable accounts annually. For some individuals, this makes up a significant amount of their retirement savings.
They did account for taxable accounts (partly by only using data for doctors with taxable accounts at Fidelity and then extrapolating from there.)
1) I wonder how many of the physicians not hitting the $17,500 limit is due to rounding error? In cases where there is a per paycheck employer match, you lose out on match if you hit the max before the last paycheck of the year. Many of these doctors may be contributing $17,450 or some trivial amount less than the limit. Generally you have to pick a whole number % to contribute (which you can change during the year). If you get any variable pay other than a straight salary, it complicates things further. I think it would be nice if we could pick a fixed $ per paycheck but none of the plans at any of the employers that my husband and I have ever worked for have that option.
2) The idea that you need a certain % of your salary in retirement is misguided. Everyone should calculate what their expenses will be and plan accordingly. In the last 4 years our household income has gone from about $300K to $425K. Does that mean we need 40% more in retirement? I assure you it does not.
Reading this on the mobile version (android galaxy s4); the first 2 ads cause the text to appear only not eh far left of the screen 1-2 words at a time for an entire paragraph. Not sure if this is a known issue, but I saved a picture if you would like to see it.
PS. Great article 😉
Thanks for the heads up. I’m trying to get that issue corrected. I think it is only with a few of the ads now. I’ll look at this post and see if I can figure out the problem.
” I can’t think of a way that makes sense unless you’re working past 70 and have RMDs.” Exactly. Why assume that everyone has retired by 70.5? Many docs work past this point. By the same token, I would not equate a doc in the early 60’s as 6 years from retirement. If you plan to keep working well into 70’s, then you should plan for an asset allocation that reflects this.
If I understand correctly, the study assumed that physicians who had retirement money and taxable accounts at Fidelity had ALL their retirement money and ALL their taxable accounts at this one firm. The former, while possible, seems unlikely, and the latter would be bizarrely irresponsible. We have our taxable and retirement funds distributed across multiple providers, although we do have both types of accounts at two different companies. The total at any one company would not be a reasonable estimate of total assets.
This also assumes that all assets that one might have are the sort that could be held at Fidelity? What about real estate? Ownership interest in a practice or other business? Collectibles?
I would assume that the holdings at Fidelity represent a small subset of total assets of these physicians. Therefore one could not draw conclusions about how much they are saving or how risky their portfolios might be.
Non physician specific Vanguard analysis from the Finance Buff:
http://thefinancebuff.com/insights-from-how-america-saves-2014.html
Link to Vanguard data is at the bottom of the post.