Welcome to the 10th episode of our Case Study Series! Today’s case study is for Mr. and Mrs. Shirts. They run their own blog Stop Ironing Shirts and I encourage everyone to head over and check out their outstanding work. Mr. Shirts and his wife face a dilemma; they have already amassed a pretty impressive nest egg, probably large enough to retire later this year. But the temptation to work a little longer to cash in that next financial milestone around the corner (bonus, vesting date, etc.) is a pretty strong incentive to stay onboard for just a little bit longer. Otherwise known as the One More Year Syndrome. In fact, in the Shirt’s case, it’s only nine months (June 2018 vs. March 2019). So, what are the tradeoffs, what are the pros and cons of retiring in 2018 vs 2019? Let’s look at the details…
Mr. and Mrs. Shirt’s situation:
Both Mr. and Mrs. Shirts will turn 36 in a few months. Some more background in Mr. Shirt’s words:
The core of my debate is I’d really hate to work full time for an extra nine months, especially if I don’t enjoy my job, then turn around and make enough money in early retirement/second career to have wasted that time in my life. I enjoy the “core” of what I do, but I dislike the hours and am working for a megacorp that’s dying in mediocrity. Not dying in a sense of bankruptcy/pension risk, but dying in terms of shareholder return, innovation, and its like watching a car wreck happen in front of you without being able to do anything.
My wife also experienced a scary injury (spinal CSF leak) that cost us almost a year of our healthy life and goals and there is a risk of this happening again. She’s most of the way through her recovery, but its given me a stronger sense of “do I really want to contribute another eight months of my life, the healthiest months of my life, to a megacorp for gigantic stacks of cash”, especially when I might make some $ in retirement. I also feel more fortunate than most because between social security, a traditional pension, and the ability to just move if a high housing cost area doesn’t work out, I am at less risk of starving in retirement than the average person.
I could probably “pack it in” and check out mentally and make it until the 2019 date, but I run a team that I care about. Its very difficult to do this when I’m not wired for it. I also wouldn’t mind supporting some charitable ventures with the additional money, especially as it relates to medical issues.
Wow, thanks for your honesty! If work is no longer fun then nine more months can be pretty painful. We’ll look into the tradeoffs below. Where do we stand with the net worth? Here are the numbers as of 2017 year-end:
After-Tax $253,856
IRAs $830,551
HSA $59,501
Deferred Comp $163,551
Home Equity $175,234 (we’ll count this in the NW, more details below)
Other (liquid/cash) $36,651
Total: $1,519,344.00
I ascertained that the Deferred Compensation package is still pre-tax. Specifically, upon retirement, the deferred comp will be paid out over the next 15 years, and each installment is still subject to income taxes (both federal and state), but not subject to payroll taxes. A little bit like a 401k plan with a forced liquidation over 15 years – and no penalty for withdrawals before age 59.5!
How about additional contributions between now and the two alternative retirement dates?
If I work until June of 2018, contributions to the accounts will be as follows: 401k: $25,700, Taxable: $37,200 (mainly RSUs vesting which I immediately sell and invest), Deferred Comp: $47,900. (Total: $110,800)If I work until March of 2019 contributions look like this, which is inclusive of the numbers referenced in the June 2018 example above: 401k: $52,300, Taxable: $103,820, Deferred Comp: $121,700. (Total: $277,820)
OK, I can see why you’re troubled! Just nine more months of work and you can cash in another $167k. You’d be leaving more than $18k a month on the table!
We don’t know our final destination at retirement. The two top candidates have very different costs:
- Eastern Appalachian Mountains (North Carolina): I’d like to run the estimate on a Home Price of $300,000 or rent of $1,750.
- Hawaii: Home Price of $750,000 or rent of $3,250/mo
It is highly unlikely we would buy our retirement home while working, which throws a wrench in the ability to finance it without employment. We could buy the house in the Appalachian option, but would be renters with Hawaii.
Nice! I think that owning a house in early retirement is probably the best option to hedge against rental inflation and Sequence Risk. For this exercise, I will work under the assumption of owning a home North Carolina. That’s because – sorry to break the news to you – but Hawaii seems too far out of reach, both as a homeowner and as a renter. You might want to consider staying in NC for a while and maybe budget for a vacation rental every year or two!
Retirement Budget
Our “core” expenses haven’t really ever exceeded $25,000/year. I define core expenses as everything except housing and healthcare.
That’s pretty low! Great job! I will budget an additional $20k in annual expenses for healthcare and housing expenses. About $10k each for healthcare (premium and out-of-pocket) and housing: property taxes, maintenance, repairs, etc. We will see below that your taxable income, even when doing Roth conversions, should stay below the Obamacare subsidy cliff, in your case $64,080 for a two-person household in 2018 (source: NerdWallet), hence, the relatively low health care expenses
How about supplemental income:
I don’t currently have a side hustle. My employment strictly prohibits that. This is something that “you don’t know until you know”, but I feel fairly confident I will earn something in retirement. Project work for clients, director’s fees, I’m a complete business nerd and am well connected in two large cities. My wife is a veterinarian by trade and maintains her licenses, it is very easy for her to pick up a half-day Saturday for $200-$350/day for extra money.
That’s nice to know that you’re flexible in retirement. Not everybody wants to chip in during retirement. I will assume that in the 2018 retirement scenario you will generate a small supplemental income. That will not be necessary for the 2019 retirement scenario.
How about Social Security and pensions?
I ascertained from Mr. S that he’s eligible for about $19,900 in annual Social Security at age 67 as of 12/31/2017. That figure goes up to $21,000 and 22,200 when working in 2018 and 2019, respectively, and putting in another year or two worth of Social Security contributions up to the cap.
I had only approximate numbers for Mrs. Shirts but I assume that she is eligible for $10,800 of her benefits. If Mr. Shirts works until 2019 then half of his Social Security is actually larger Mrs. Shirt’s own benefit. Also, given that you’re both still pretty young I’d like to apply a 25% haircut to your benefits to account for future benefit cuts:
Safe Withdrawal Rate Calculations
I will run the numbers twice, once for a June 2018 retirement date and once for March 2019 retirement with the additional savings. Both calculations require some assumptions about capital market returns between now and then, so I’ll try to be cautious and conservative with the capital market projections.
Retire in 2018:
The portfolio grows to just under $1.7m. Net of the purchase price for the house, that leaves about $1.375m in today’s dollars.
Given that the asset position is a little bit thin and you indicated that you’re open to generate some supplemental income, let’s assume you bring in another $1,000 per month for five years, starting in 2019 (when you’re in a lower tax bracket), adjusted for inflation. I found the $1,000 figure will generate a 4% SWR and it’s also easily attainable with a few side gigs here and there, especially with Mrs. Shirt’s weekend vet gigs. You can go to the spreadsheet and play around with the numbers and see how the SWRs change.
Some additional assumptions:
- 70/30 asset allocation. I usually find that 80/20 is ideal in a bare-bones scenario without any additional flows (pensions, SocSec). 60/40 is ideal for older young retirees who expect sizable supplemental flows a few years down the road. You are smack in the middle and I found that 70/30 works best.
- 60-year/720-month horizon.
- 0% final value target
The Google Sheet is the usual setup as I explained in Part 7 of the SWR series:
Google Sheet with simulation results (2018)
If you like to make changes, please first save your own copy!!!
Let’s check out the results:
The Failsafe is 3.88% (occurred during the 1965/66 retirement cohorts, exposed to the 1970s/early 80s mess). A 4% WR would have generated a 5% failure rate. Pretty good! Overall, this is not too surprising. The non-COLA pension and Social Security are 18 and 31 years away, respectively, too late to compensate for a drawdown early on, but with the supplemental income, you get to your desired 4% SWR. Relative to your $1,375,000 initial portfolio, that’s $55k of annual cash flow. Budget about $5k for taxes and you can spend around $50k per year. Even the failsafe 3.88% rate would leave you with $53k, probably $48k after taxes.
So, just from the backtesting/simulation perspective your June 2018 is a “go” though there are some wrinkles as I will detail later!
Retire in 2019:
With the additional contributions and some moderate capital returns (assuming 4% nominal over the 9 months for equities), we now got about $1,580,000 in the portfolio. This is already after paying for the house!
Google Sheet with simulation results (2019)
I maintain the 70/30 allocation assumption. Also, the “stickler” that I am, I set the horizon to 711 months, 9 months shorter than in the other simulation just to be consistent and comparable to the other scenario! The Results:
Now we have a slightly lower safe withdrawal rate, but with the additional financial net worth ($1,581k), those lower rates still generate about $58,500 p.a. for the failsafe and about $60,500 for the 3.83% WR. That’s significantly more. After tax, you’re now looking at a sustainable consumption of about $53k-$55k p.a., depending on how “safe” you want to be. It’s about a $5k cushion per year over and above the 2018 retirement scenario, every year, in a 50-60 year retirement. For 9 more months of work.
OK, I’ve talked to people who hate their work but I’ve never met anyone who’d walk away from a 10% boost to the post-retirement consumption for nine months of work. Check out the Grumpus Maximus blog. He faced the problem of grinding through not months but years of a tough job for that “Golden Albatross” of a nice government pension. Incidentally, Grumpus was the guest on the ChooseFI podcast episode 57 on Monday!
Cash Flow Simulations
As always, I’m interested in how the withdrawal plan would work with the actual portfolio with all the different accounts spread over taxable, tax-deferred and tax-free accounts. I will do the analysis for the 2018 retirement date using the following assumptions:
- All variables are nominal. Tax brackets, consumption targets, O-care cliff etc. all increase by my assumed 2% inflation rate per year.
- After you sold your house (that’s where the big cash position comes from), you will use $200k of your cash, $120k of taxable investments and the first installment of the deferred comp (1/15th of the account balance) for the home purchase and the rest of the 2018 CY living expenses.
- You keep a 60/40 allocation in the deferred plan and the 401k and a 100/0 allocation in the taxable and Roth accounts.
- Your state taxes are 5.75% in the state of North Carolina.
- The mandatory Deferred Comp plan withdrawals are 1/15th the first year, 1/14th the second year, 1/13th the third year, and so on, until the remainder is withdrawn in the 15th year.
- Capital Market returns are my conservative estimates, as usual in my case studies. Only 5.75% for equity returns going forward!
- You withdraw from the taxable account until it’s depleted. Scary thought, I know, but that’s where the Roth Ladder comes in!
- I lump together your existing HSA with the new Roth. Make sure you keep all of your out-of-pocket receipts from now on so you can access the HSA balance later in retirement if you need to! The withdrawal doesn’t have to occur in the same year as the health expenditure!
- Starting in 2019 you perform a Roth Conversion ladder, which shows up in the tables as a withdrawal from the 401(k) and a negative withdrawal (=contribution) into the Roth.
- After you run out of money in the taxable account you’ll continue the Roth conversions but also start taking out money you Roth-converted earlier. You should have enough principal from conversions 5+ years ago, plus your HSA which you can use to fund health expenditures (make sure you keep all the receipts from now on!). Just like in the case study from a few weeks ago, you now still make a net contribution into the Roth, but you also take out a gross sum. For example, in 2026, you move $46,087 from the 401(k) to the Roth, but you take out $20,365 (=46,087-$25,722) from the Roth (principal from 5+ years before) to make ends meet!
- In 2041, the year you turn 59.5, you can now use the 401k. You live off that until it runs out (in reality it may not run out at all, of course) and let the Roth grow. If you do run out of the 401k you’ll use your Roth. Here in the simple simulation that happens in 2049.
- A very important comment about the Roth Conversions: Make sure you keep your Modified Adjusted Gross Income (MAGI) below the threshold! I assume here you do the conversion up to $1,000 below the threshold every year. Just to have a cushion! See Part 3 of this table for the evolution of the Obamacare Cliff and your personal MAGI. Your income is everything: interest, capital gains, supplmental income, Roth conversiond. Withdrawing cost basis from the taxable account doesn’t count, of course.
- Your pension starts in the year 2037. Eight monthly installments that year, then $13,536 after that.
- I assume that you both take Social Security in the year 2049 when you turn 67. You should, of course, revisit the benefit timing once more when you get closer. Married couples can do some benefit hacking that will give higher lifetime expected Social Security payments than the actuarially fair value!
Results:
The plan seems to work. You even grow your real portfolio value from $1.3m in 2019 to about $1.5m by age 70. By that time, you’d have shifted your 401k to the Roth and you should have no issues with Required Minimum Distributions. In fact, after age 67, your portfolio will keep growing because Social Security kicks in. The plan also generates a natural glidepath to 100% equities. If you’re not comfortable with that you can of course shift to more bonds inside your Roth once the Roth becomes the primary retirement account. If for some reason the Roth conversion ladder doesn’t deliver enough cash flow (remember, only the principal from 5+ years ago is fair game), you might consider the 72(t)/SEPP route as a last resort!
Conclusion
I think you’re likely ready to retire either way. Of course, one concern is your wife’s health. Another costly health episode could lower your combined supplemental income potential and also bust your budget in early retirement. Two bad risks with a correlation of one! This risk alone would entice me to just delay retirement by another short 9 months.
The other reason to hold off retirement just a little bit longer is the housing situation. Retiring in 2018 would seriously deplete your taxable savings and you’d have to really cross your fingers that the Roth Conversion Ladder works the way you planned. But then again, you can always use the 72(t)/SEPP route to tap the 401k savings before age 59.5.
Another unpredictable risk is the survival of the Obamacare subsidies. I’m pretty confident that some form of O-care will be permanent, but the subsidies could easily become a lot less generous over time. As we saw with the GOP tax change, politicians can take away “tax hacks” like the state and local tax deduction with a vote in each chamber and a president’s signature. I would not tie my retirement to the survival of those healthcare subsidies! Just look at the mortgage interest deduction which has become essentially useless for most taxpayers due to the high standard deduction. In the next tax reform, this deduction will go out the window forever because so few people will miss it then. Who would have thought ten years ago that the formerly sacred cow of the mortgage deduction can ever go away?
So, whatever retirement date you pick, best of luck! Keep us updated!