Ask Big Ern: A Safe Withdrawal Rate Case Study for “Rene”

Welcome! Today is the third installment of our Case Study Series. Please check out the other two posts here if you haven’t done so already:

Today’s volunteer “Rene” (not her real name) was laid off earlier in 2017 and is now living off her severance package. She wonders if she has enough of a nest egg to simply call it quits and retire in her late 40s. And many other questions: if/how/when to annuitize any of her assets and what accounts to draw down first? So many questions! As I pointed out in Part 17 of the Safe Withdrawal Series, a safe withdrawal rate calculation has to be a highly customized affair and that’s what we’ll do today again. Let’s see what the numbers say!

Rene’s situation

Hello ERN!

I would love to be a case study! My job was unexpectedly eliminated in Jan 2017. Some days I still can’t believe it – still waking up feeling devastated six months later. Thankfully, I received 12 months of severance (100% pay for 6 mos., 60% pay for 6 months).

Oh, no, I’m sorry to hear that! But consider yourself lucky to have such a generous severance package! That could make a big difference!

Age: 48 (will be 49 when severance ends in January 2018). My husband of 24 years (age 60) and I maintain separate finances. Here are my numbers.

Investments: Total ($923,074)

  • IRA: $304,270
  • Taxable: $219,305
  • Roth IRA: $28,355
  • 457(b): $160,962
  • HSA: $21,848
  • Savings: $78,255 (+ $17,000 additional savings from severance through 12/31/17)
  • Frozen “Pension” Plan: $93,079 (cash balance)
  • Asset allocation is 52/40/8 across all accounts. The three largest accounts are in low cost index funds at Vanguard. Online savings account – 1.15% APY.

First of all, congrats on that impressive portfolio. I also love the idea of saving from the remainder of the severance package payments to stash away some more cash before next year!

Liabilities: 2016 vehicle loan (0% interest for 36 mos.) – will be paid off using final January 2018 severance check and 26 weeks of unemployment checks in 2018.

Home: $300,000 equity, no mortgage – no set plans to relocate or downsize. Live in a MCOL area (Ohio).

Medical/Dental: Husband’s retiree health insurance with reasonable monthly premium ($245) deductibles and coinsurance.

Social Security: If I never work again, my payments are estimated to be:

  • $2,040 at age 67 (FRA) (or $1,570 if only receive 77%)
  • $2,549 at age 70 (or $,1962 if only receive 77%)

I would probably give your Social Security less of a haircut than 23%. At your current age 48, you have only 7 more years to make it to 55, which I always consider the safe age beyond which no politician wants to mess with anybody’s Social Security. In my calculations here I will only apply a 10% “haircut”

Projected First Year of Retirement Annual Expenses – $32,000  (tracked every penny since Jan. 2014):

  • Non-Discretionary ($18,840) – Property Taxes/Utilities/Food/Medical Premiums/Home Repairs
  • Discretionary ($10,140) – Gifts/Grooming/Car-related/Clothes/Vacation/Entertainment
  • add’l cushion/other ($3,020)

Awesome job! I love how you have such a good idea about your retirement budget already by keeping detailed records on your expenses. That makes the whole calculation so much easier!

Additional Information:

  • 457b has to be depleted within 10 years of last severance check (i.e. Jan 2028). Nongovernmental rules prohibit IRA rollover. May take lump sum or installments over a maximum 10 year period (2018-2028). The irrevocable decision must be made in Jan. 2018.
  • The cash balance pension plan is accessible at age 55. May rollover to IRA or leave at employer earning treasury bill rates.
  • No biological children. Leaving an inheritance or money to spouse is not important. (Spouse has traditional pension, paid off rental property, an annuity, and other cash assets).
  • Tax preparer runs both MFJ and MFS. We usually file MFJ with each paying proportionate taxes owed, if any.
  • My husband retired in Feb. 2017. For 2018 he will have annual income of roughly $55,000  ($31,200 pension + $12,000 current part time job + $11,653 rental income, after expenses). In 2019 (age 62) he plans to drop the part-time job, take SS for an annual income of roughly $65,000 ($31,200 pension + $22,440 SS + $11,653 rental income).

Your husband’s and your combined adjusted gross income will be in the mid to high $80,000s (depending on how much you withdraw from principal vs. capital gains), which will be just enough to stay in the 15% federal tax bracket. That will also ensure that your taxable qualified dividend income and your long-term dividends will be taxed at a 0% rate on your federal return. That’s great news!

Questions:

  • If I remain unable to locate a position, can I retire after severance ends in Jan 2018?
  • If yes, how should I draw down my funds? Start SS at age 67 or 70.
  • Should I annuitize any part of my funds? If yes, now or later?
  • Am I missing anything?

All great questions. Let’s look at the numbers in more detail.

The Safe Withdrawal Rate Calculation

Let’s plug the numbers into the Google Spreadsheet. For your initial net worth, I use only your current financial assets and exclude your pension. Rather, I model the pension as a one-time cash flow into your investment portfolio after 6 years. Social Security (after a moderate haircut and accounting for taxes) adds another roughly quarter percent per month after 21 years. That’s worth almost 3% of your portfolio in today’s dollars and will help fund 2/3 of your retirement once you reach age 70!

Case Study Rene Table03
SWR parameters.

The Google Sheet with all the simulation results (see SWR Series part 7 for more details):

SWR Case Study for Rene

Thanks to a relative “late early retirement” you are much closer to receiving Social Security than most FIRE planners. You can jack up the initial SWR to probably around 4.60% (relative to the non-pension net worth), even with today’s elevated Shiller CAPE (which now very, very slightly exceeds 30, but I’ll still look at the 20-30 range):

Case Study Rene Table04
Main results from the Google Sheet. Even with the elevated CAPE ratio, Rene can withdraw well over 4%, thanks to the pension and generous Social Security benefits!

This would imply about $38,000 per year (0.046 times $829,995), significantly more than your consumption target, even when accounting for taxes. More details on the tax issue below.

So, from the pure SWR simulation side, this looks like successful FIRE plan!

A Cash Flow Analysis

Another dimension of the withdrawal strategy deals with whether the withdrawal strategy generates enough income before you are eligible to tap your IRA penalty-free. Here are my assumptions:

  • Account balances don’t grow between now and December 31. (but they don’t shrink either, knock on wood)
  • I suggest you keep a 60/40 asset allocation in your IRA, Roth and HSA accounts. 100% equities in the taxable account and 30% equities, 70% bonds in the 457b. Combined with your savings account and the pension, this creates an initial equity weight of about 52%, but you increase this over time to about 60% after the pension moves to the IRA. Why keep such a lopsided allocation? It’s more efficient to keep equities in the taxable account and bonds in the tax-advantaged vehicles.
  • If you are completely uncomfortable with this allocation, just set everything to 65/35 in Taxable, 457b, IRA, HSA, and Roth. Together with the savings account you will also just about reach a 60/30/10 Stock/Bond/Cash allocation, which is what you seem to be comfortable with. But the more lopsided allocation comes out slightly ahead.
  • For capital market returns, I use the 5.75% equity expected return for the first ten years (see last week’s post), then 6.5% for another 5 years and 7% after that. These are all nominal returns. Bond returns will slowly rise to 3.25% and cash returns (i.e., money market, T-bills) will slowly increase to 2.25%. CPI is assumed at 2% p.a. throughout. Actual returns may exceed this or not. But I intentionally picked conservative numbers, just to be on the safe side.
  • Do not take the 457b as a lump-sum! This would push your combined taxable income into the federal (and also state) marginal tax stratosphere. You want to draw down your 457b gradually over the 10 years. I assume about $17k in 2018 and then adjusted for 2% inflation every year. I calibrated the initial withdrawal to exactly deplete the account by 2028. But the actual numbers will differ depending on realized market returns, of course.
  • You slowly draw down your savings account to the 2x annual consumption target, after which you only withdraw the annual interest income.
  • Also, notice that the total withdrawals are a little bit higher than your annual consumption target because you have to withdraw enough to also cover the tax bill! In fact, I solved for the amounts withdrawn from the Taxable account to make sure the after-tax withdrawal exactly matches your consumption target. I used the Excel Solver function for that.
  • I model the Pension to IRA rollover in 2024 as a (tax-neutral) withdrawal from the pension and “negative withdrawal” from the IRA.
  • I assume that your ordinary income is taxed at 12%. That’s less than the marginal tax rate (presumably 15% federal, 4% for state taxes in Ohio, according to the Tax Foundation), but it has to do with spreading out the average tax liability proportionately between your and your husband’s separate finances.
  • Because part of your withdrawals in your taxable account comes from a) dividends b) long-term capital gains and c) cost basis, all of which are taxed at a zero federal rate and a+b are taxed at a relatively low state tax, I assume that you owe roughly 5% of the withdrawals in taxes. The exact tax calculation might vary but I think this will be pretty close. Strictly speaking, you should assign a tax liability only for state taxes and only for capital gains and dividends (not cost basis withdrawals) when you assign tax bill between you and your husband, so no more than 4% but I budget 5% to be on the safe side.
  • At age 60, when you can withdraw from your retirement accounts, I assume you withdraw from all investment accounts proportional to their respective size. I also assume you do the same for the HSA. Your tax burden increases by a little bit because more money comes out of the taxable IRA.

With these calculations I want to check for two potential problems:

  • Do you have enough money to withdraw in 2028, the last year before you can make penalty-free withdrawals from the retirement accounts?
  • Do you run out of money before Social Security kicks in, in the year 2039?

Here are the results, spread over two tables because the table got too wide. So in the first table I have the asset levels at the beginning of the year and the % asset allocation and in the second table we have the withdrawals, the tax calculations, and the return assumptions:

  • At the beginning of 2028, you still have more than $244k in savings and the taxable brokerage account (in addition to the roughly $39k HSA that can be tapped at any time for qualified expenses). So, you will have more than sufficient assets at age 59. That’s good news, so you don’t run into the issue of having to dig into retirement accounts and face the 10% penalty.
  • Despite drawing down your investments for another decade, you still have sufficient assets at age 70 when Social Security kicks in. You still have roughly $600k left in today’s dollars. And remember, that’s when Social Security provides about 3/4 of your living expenses!
Case Study Rene Table01
Cash Flow Table, part 1.
Case Study Rene Table02
Cash Flow Table, part 2.

So, even from the cash flow analysis, with relatively conservative assumptions about expected returns and factoring in taxes, your plan looks like it’s going to work!

A Roth Conversion Ladder?

Your case doesn’t look like a good candidate for the “never pay taxes again” Roth Conversion Ladder. That’s because you and your husband will have too much ordinary income already to fill up not just the “tax-free bracket” (standard deduction plus exemptions) but even the 10% bracket and a good chunk of the 15% bracket.

But it’s not a reason to totally ignore the Roth Conversion Ladder. Notice that at age 70, you’ll have significant Social Security income and also make further (minimum required!) distributions from your taxable IRA. This might push you beyond the 15% tax bracket. One way to mitigate that risk is to max out the 15% tax bracket starting in 2018. Simply do the following:

  • Wait until December to get a better picture of your YTD taxable income and check on the web what the tax brackets are for that year (on taxfoundation.org they usually post the updated brackets in the Fall of each year).
  • Calculate how much income you can generate to stay within the 15% bracket: Standard Deduction + 2 Exemptions plus the upper end of 15% bracket for Married filing jointly. In 2017, this would be $12,700 + 2 times $4,050 + $75,900 = $97,700, according to the Tax Foundation, but this will be adjusted every year to account for inflation.
  • Calculate the sum of income that will make it into your adjusted gross income (AGI): Dividends, interest, capital gains, 0.85 times Social Security, rental income, part time job, 457b distributions. Exclude the withdrawal of cost basis in the taxable brokerage account and principal withdrawals from your savings account in that number!
  • Calculate $97,700 minus your projected AGI and do a conversion of that amount from the IRA to the Roth. This won’t be much, maybe $10-20k per year.

This conversion will cost you 15% marginal federal tax plus 4% state tax but it will save you that much and potentially more in the future. This is insurance against slipping into higher brackets in the future or, equally likely, increases in federal and state tax rates.

Here are my thoughts on annuities:

  • You have an awesome annuity: Social Security. It’s adjusted for inflation (in contrast to many commercially available products) and there are no fees and broker commissions. If you consider yourself generally healthy with a history of longevity in your family it is a good idea for a female retiree to push Social Security all the way to age 70 to get the maximum benefit from an actuarial point of view. Remember, you are trying to hedge longevity risk with this!
  • You should probably avoid transforming any of your investments into an annuity. Fees are usually too high and since you already cover a very large portion of your living expenses at age 70+, I would stay away from annuities.
  • One exception to the above rule is the company pension plan. You might want to revisit the pros and cons of taking a pension at age 55. Some corporations offer extremely good deals on annuities, from an actuarial point of view. Mine does, for example. But keep in mind that corporate plans don’t normally offer COLA (cost of living adjustments) so dollar for dollar this is worth less than Social Security, of course. You don’t want to leave the pension post age 55 to sit and earn only the 3-month T-Bill rate. So, either annuitize or roll over. Don’t leave it to accumulate after age 55!

One other idea

Since you insist on having 2 years worth of expenses you could do a little bit better by keeping a CD ladder rather than keeping the entire loot in a money market account. Most of the time, 2-year CDs yield more than the money market account. So, you could shift 1/8 of the savings account to a 2-year CD and over time you’ll have a CD coming due every quarter with exactly one quarter worth of living expenses. If the market is heavily under water you withdraw the CD, otherwise, you roll into another 2-year CD.

Conclusion

Your case looks like a good example of a successful FIRE plan. Despite a high income during your working career you never fell into the trap of lifestyle inflation. Your nest egg relative to your annual expenses is surely big enough to support gross withdrawals of $35k, $32k after tax.

But I don’t want to discourage you from looking for another job either. The longer you work and the longer you stash money into your nest egg the safer the retirement becomes. Who knows, what will happen to health expenses in the future. That’s hard to budgetfor! But you can look for a job from a position of great confidence knowing that you might as well just retire! Best of luck!

29 thoughts on “Ask Big Ern: A Safe Withdrawal Rate Case Study for “Rene”

  1. Great analysis. Do you ever factor in the impact of a divorce when the wife has the larger nest egg? I’m assuming all the assets would be split down the middle leaving her with only half of her retirement account? Or maybe that doesn’t work that way.

    1. Wow! Very touchy issue. Rene and her husband have “separate finances” and that may or may not translate to a divorce. Good point! This is another factor that has to be looked at on a case-by-case basis!

      1. Hello, this is “Rene.” I thought you might be interested in how we came to have separate finances and how we would split assets in the event of a divorce.

        When we met (1990) I was a full-time college student earning minimum wage at a part-time on-campus job. He had been in the workforce for 15 years (since high school) and was earning $70,000+ with overtime. He was recently divorced and a little bitter over how much equity he had to give his ex out of their house. When I started working (1992) we simply kept things separate. He was pretty well-established and I knew I would be financially successful (and being a feminist I did not want him to think I was after his money:) I began out earning him in 2002.

        We’re both very frugal and prodigious savers. He, however, is extremely risk-averse. He has never invested in the stock market. (I could not convince him to start a 401(k) when his employer began offering it).

        For the last 24 years a percentage of our salaries went into my checking account. I pay the household bills from this account. We negotiate big ticket items. E.g., two years ago I had the windows replaced. He bought new gutters and had the deck torn down and rebuilt.

        In the event of a divorce, we agreed he keeps his assets (2 annuities (one inherited), pension, CDs, cash, rental property) and I keep mine. We would have to agree on how to equitably split the equity in the house. He put up a considerable down payment, but I’ve paid more towards the house because I insisted on moving and I made more money. Although his previous home was paid off, and he would have been content there, I hated the layout, the (small) size and the neighborhood. And let’s be honest, he bought it with the first wife.

        It’s strange to some, but it has worked really well for us.

        1. Great plan, and I hope you didn’t think I was predicting the worst. I’ve been married 39 years and counting and wouldn’t do that. It was more of an intellectual curiosity about how people should think about retirement and whether that was even a consideration. You are a remarkable person Rene, and it was a real pleasure observing you and Big Ern interacting. Thanks for sharing so much of your life, it really helps the community to learn from others.

  2. Love reading this case studies ERN. They always draw out some relevance to our own plan even though circumstances are unique. Learning a ton through these posts.

    Case (forgive the pun) in point is the following issue that we have been going back and forward on.
    It relates to asset allocation across taxable and tax-deferred accounts and the point you mentioned about tax-efficiency of bonds in tax-advantaged accounts.

    Our plan is to hold a mix of equities/bonds in our taxable as it is that account we will be drawing from for first 8.5 years until I hit age 59.5. In taxable that ratio will be 60:40. In tax-deferred across both our 401K rollovers, we plan to hold mainly equities (90:10 ratio). With an overall portfolio target allocation of 75:25 equities to bonds. If market is down significantly, we will of course be selling bonds from the taxable account rather than deflated equities. Thus important for us to have sufficient bond ballast in that account if markets are depressed over first 5-7 yrs. SoR and all that.

    Your point on tax efficiency of bonds being in tax-deferred reminds me of a long thread on Bogleheads I read a few months ago and our specific issue of whether we should just spread our asset allocation evenly across taxable and tax-deferred. ie. 75:25 across the board. Thoughts on whether it is critical to adjust our asset allocation plan or is it something that may not markedly change our overall portfolio growth?

    Best,

    Mr. PIE.

    1. Thanks, Dr. PIE!
      You bring up a very important topic: Asset Location (as opposed to asset ALlocation). This topic is so important that it deserves its own blog post and it’s on my to-do list. Stay tuned!!!
      Briefly, here’s where I’m coming from: Take a simple example (not precisely Rene’s case, but qualitatively similar):
      Taxable account $100. All cost basis, no capital gains
      Tax-deferred account (401k): $125.
      The marginal tax on withdrawals from the 401k is 20%. So today’s after-tax value is $200 = 100+125*0.8.
      The marginal tax on interest income is also 20%, and since we’re in the first two tax brackets the marginal tax on dividends/long-term capital gains is 0%.
      Bond interest = bond return = 2%. No capital gains/losses from bonds.
      Equity dividend yield plus capital gains = 5%

      Take 2 extreme cases:
      Case 1: All equities in taxable, all bonds in 401k. This is a 50/50 allocation in today’s after-tax dollars. After 1 year we have $105 in the taxable account, all tax-free and $127.50 in the 401k, which is $102 after tax. Total $207.00
      Case 2: All bonds in taxable, all equities in 401k. Same 50/50 allocation in today’s after-tax dollars. After 1 year we have $102 in the taxable account, which is 101.60 after tax. And we have $131.25 in the 401k, which is $105.00 after tax. Total $206.60 after tax.

      So, equities in the taxable account generate higher average returns. It’s because equity returns are taxed at a higher rate in the 401k than in taxable, while bond returns are taxed the same either way.

      There are at least two additional advantages of equities in the taxable account:
      1: tax loss harvesting. Since equities are more volatile there are more opportunities to use tax loss harvesting
      2: In the above advantage the adjusted gross income (AGI) is higher in case 2. Thus, you have less room to max out the 15% bracket for the Roth conversion.

      Best,
      Big ERN

  3. Thank you for the quick and detailed response. This explanation is very helpful indeed and I wish I had it before I was sucked down the Bogleheads forum rabbit hole! The reminders of TLH and Roth conversion advantages are great also.

    Look forward to your asset location post. I am sure there are many like us who are asking similar type of questions.

    We are working on a little post on our approach to taming the SoR beast. If it can be tamed! Look out for that next Monday and feel free to critique, bash and let us know if we are in cloud-cuckoo land!

    Best,

    Dr. PIE.

  4. Many thanks for doing these case studies! Nothing is as educational as applying a theory to a real life situation.

    Something I may have missed is why the failure rate for CAPE>30 regime is ever below the other two. Does everything seem correct there?

    1. Good question! Only around in the late 1990 to 2000 did the CAPE go above 30. If you had started with a 4.6% SWR back then you would have very seriously depleted your portfolio by now but since you’re so close to drawing Social Security you will likely make it. So, as crazy as it sounds, the 2000s were not the worst period for her plan. A 1965 or 1966 starting date with merely elevated CAPE was the worst case scenario.

  5. Big ERN,
    Suppose you don’t need all of the money that the IRS requires you withdrawal from your taxable IRA at age 70-1/2. Would it be best to convert this to a Roth prior to that age (throughout your 60’s) in order to avoid the requirement, or would it be equally as good to reinvest the “extra” in a non-IRA account?

    1. In my view, it all depends on the marginal tax landscape. I would never try to do the conversion at a higher marginal tax rate today than what I expect in the future. But if you do a piecemeal conversion every year, converting to max out the tax bracket? That would be a good idea!

  6. Love these case studies. Even though FI is a ways out this helps me think and plan for that hopeful inevitability.

  7. Big ERN,
    The article below about the CAPE 10 made me think about your spreadsheet tool since you rely on it there. My takeaway is that people should use the CAPE as a general indicator but don’t blindly trust it. It looks like the market is slightly overvalued but I’m not going to change my investment strategy because the CAPE is “high”. I will proceed with caution.

    http://www.etf.com/sections/index-investor-corner/swedroe-metric-dire-need-context?nopaging=1

    1. Thanks!
      First, we have to make a distinction. Do we use the CAPE to time the stock/bond allocation? Bad idea. We seem to agree on that one.
      Or do we use the CAPE to gauge the appropriate SWR? Not a bad idea, I would argue.

      I agree that today’s high CAPE is much less of a problem than some naysayers want to make it. The guys at GMO and Hussman, mentioned in that article, are a running joke in the industry. So,I’m the first to admit that a CAPE of 30 in 2017 is probably comparable to a CAPE of maybe 25 in the longer history. But that’s still elevated. Hence, the slightly leaner SWRs.

      What I’m trying to say is that for every crazy crackpot on the on the one side (GMO) there is also a finance buffoon who ignores valuation altogether. Both extreme views will not do well in the long-term.

      Cheers!

  8. Another awesome post! Love these case studies. I would love to submit my info to you for assessment/publication!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.