Ask Big Ern: A Safe Withdrawal Rate Case Study for Mrs. “Wish I Could Surf”

earlyretirementnow.com

7 years ago

Welcome to a new Case Study! This time, Mrs. “Wish I Could Surf” (not her real name) volunteered to open the doors to her finances. And every case study brings up something new to learn for yours truly. Today’s challenge: How would “alternative” investments factor into the Safe Withdrawal Rate exercise? Peer Street, Hard Money Lenders, Lendingclub, Prosper, etc. have gained a lot of popularity, especially in the FIRE crowd. When calculating safe withdrawal rates, I have only worked with stock/bond/cash portfolios because they are the asset classes with returns going back 100+ years. Doing the SWR exercise for a portfolio of Peer Street loans will require some “hacking” in my Safe Withdrawal Rate Google Sheet!

Further challenges come from the fact that Mrs. and Mr. Surf keep their finances separate (similar situation as in the Case Study for Rene) and Mr. Surf will still be working for a number of years, so we have to make some assumptions on how to assign the tax burden between Mr. and Mrs. Surf. Lots of work to do! So let’s get started and look at Mrs. Surf’s finances…

Mrs. Surf’s situation:

I’m 44 years old and the goal is for the money to last 50 years or so.

I also have a little bit of a unique situation in that I was divorced 9 years ago and remarried last summer. My husband/partner and I choose to run our finances separately – he still works and likely will for the next 7-10 years though (maybe less, just depends). I’ll be able to access health care through his company for my daughter and me for that time period.

Though we run our finances separately we split the cost of the mortgage and property taxes on the house I own.

Mortgage has slightly less than $260,000 outstanding. It’s a 3.279% fixed 30 year rate and I’ve paid ahead from time to time. I’m taking your advice now and will only pay the minimum… 🙂 I think I have about 14 years left until that’s paid off completely. I pay $1,670/month on the mortgage.

Social Security: the benefits are about $2,130/month at age 67 (in 23 years).

Actually, I calculated that the mortgage has about 16 years left so I will use that number. How about the current portfolio?

Hard Money Lending Private Equity $100,000

Checking $5,000

Savings $7,500

Betterment $269,177

Betterment Preserve Capital $6,019

401K $224,670

Lending Club – $933

Vanguard Brokerage $272,892

Vanguard Roth IRA $36,153

Vanguard Rollover 401K $273,514

Peer Street $415,778

Total $1,611,636

Very impressive numbers! Mrs. Surf is also expecting a deferred compensation package in 2018 worth around $80,000. That’s a nice tax arbitrage: she was able to shift some of her 2017 income into 2018 when she’s likely going to enjoy a slightly lower marginal rate! Even more important: She can double-dip in the 401k and max out another $18,000 in pre-tax contributions for the next calendar year! Maybe even $18,500, if the IRA raises the limit in 2018. Together with an expected $16k tax refund next year her finances in early 2018 are projected to look extremely solid:

Cash/MM $12,500
Brokerage $591,488 (taxable accounts at Vanguard plus Betterment plus the after-tax deferred compensation package).
401k/Rollover401k $516,184 (existing 401k plus rollover IRA plus $18k contribution in early 2018)
Roth $36,153
Private Equity $100,000
PS/LC $416,711 (I will lump together Peer Street and the tiny LendingClub balance)
Total $1,673,036

How about expenses?

The $4,000 monthly budget is all-in for my daughter and me. Includes my half of the mortgage and my estimated tax liability. The $4,000 is honestly a higher budget than what I actually spend but if the shit hit the fan and $4,000 assumes I pay the entire mortgage and property taxes. My actual spend is closer to $2,500. I modeled my $4,000 off if something happened to my husband and we didn’t have his income and I paid the entire mortgage.

OK, we’re going to work with a $48,000 p.a. expense target. Roughly $10,000 of that is her share of the mortgage payment, which we won’t have to adjust for inflation and she has to pay that amount for only 16 more years (yay!). $38,000 is the actual long-term annual consumption target that needs to be inflation-adjusted!

The Safe Withdrawal Rate Analysis

Historically, Mrs. Surf’s Hard Money Loan has paid in excess of 10% in annual dividends. Peer Street about 8% p.a. The challenge for the SWR analysis is that we don’t have much in the way of historical returns for Peer Street investments. Certainly not going back to 1871! Of course, just because we don’t have historical returns doesn’t mean that it’s a good assumption for us to model Peer Street paying 8% returns every year going forward. It would be huge mistake to model the real-estate-backed loans as safe Treasury Bonds with an 8% coupon! If we were to have another 2008-style recession I’m sure the Peer Street and private hard money loans would take a big nosedive! So, here’s my proposal for how to “hack” the ERN Google Sheet for this situation:

I assume, for the historical backtests, that half of the Private Equity plus Peer Street portfolio, $258,356, is set aside and will earn a 7% (nominal) yield forever, paid in the “supplemental income” column, while the other half will behave just like an S&P500 equity index portfolio. If we average over the two components we’ll get an expected real return of just about 6%, exactly the same as your current 8% nominal Peer Street yield minus 2% inflation. But we still generate a pretty substantial business cycle exposure in the alternative investments in the simulations.
The total portfolio value that’s used to calculate the safe withdrawal amount is then $1,414,681.
I assume that the 401k accounts are 50%/50% in stocks/bonds, all other brokerage accounts are 100% equities and the small cash allocation will earn a money market return. This would imply an allocation of 81%/18%1% in stocks/bonds/cash. Pretty ideal alloction for your situation.
The monthly cash flow from the $258,356 is $1,507 (nominal). For the first 16 years, that number is netted against the $835 mortgage payment. We inflation-adjust these two payment streams because in the Google Sheet everything is supposed to be in real dollars. I assume a 2% annual inflation rate.
After 23 years you will also get Social Security, worth $2,130 in today’s dollars, net of a 15% haircut, to account for potential future benefit cuts. Just to be on the safe side!
I assume a 600 month/50 year horizon and a 50% final value target. Notice that in the Google Sheet, I set the final asset target value to 43.2%. That’s because the $258,356 set aside for the supplemental income would have depreciated after 50 years worth of inflation (2% p.a.) to about 6.8% of today’s portfolio value. We add that back to the final value.
All other assumptions are as in my baseline Google Sheet.

Google Sheets Link to the SWR Study

Given all those parameters, the failsafe withdrawal rate is 4.12%. At a 4.33% withdrawal rate, we still keep the failure rate at 5% in historical simulations. All assuming an elevated CAPE ratio! 4.12% multiplied by the $1,414,681 portfolio value (not the entire $1.6m, that would be double-counting!) is about $58,000 in annual withdrawals (though reduced by the supplemental flows). If we budget about $15,000 in taxes, you’d still have $43,000 left over, more than enough for your $38,000 consumption target. So, even with a failsafe withdrawal rate, you’ll have a cushion of about $5,000 annual expenses!

Output From Google Sheets: Safe Withdrawal Rates North of 4% even when the CAPE is high! That’s thanks to the substantial supplemental income!

Also, I should say it again: all we needed to fund with the Google Sheet calculations is the annual consumption in excess of the mortgage payment ($38k). That’s because we already pay for the mortgage ($835 a month) with some of the Peer Street supplemental income flow! So, from the SWR calculation, your FIRE plan seems extremely safe. Of course, this is all subject to the usual disclaimers, the limitations of historical returns and the modeling assumption of your real estate returns.

Portfolio balances up to age 70

As usual in these case studies, I like to see how the specific account balances evolve over time to avoid any cash flow problems, where the taxable accounts run out before the 401k and IRAs can be tapped. Here are the assumptions:

Asset Location vs. Asset Allocation: I normally recommend holding bonds in tax-deferred accounts and equities in taxable accounts. In the calculations below, I assume that the taxable accounts and Roth are 100% equities, the 401k is 50% equities and 50% bonds.
Stock and bond returns are the same as I usually assume in these case studies: 5.75%/2.00% early on, then shifting up to 7.00%/3.25% over time (nominal).
I assume that the hard money loans and Peer Street investments keep generating 7% p.a. (nominal!) and this income is skimmed off every year to pay for expenses. They will likely pay more, but let’s be conservative.
I ignore your small cash/money market allocation for this calculation.
I assume marginal income taxes are 30% for ordinary income and 23% for dividends/Long-term Capital gains in 2018 because your $80k deferred compensation pushes you into a higher bracket. Then, going forward, you’ll enjoy lower marginal rates. In fact, you will likely manage to stay inside the first two tax brackets for federal taxes, so capital gains and dividends will be taxed only at an 8% marginal tax in your state. I suspect that
I assume you keep contributing $5,500 p.a. to your Roth for as long as your husband is working (7 years minimum). Starting in the year when you turn 50 (presumably 2023), you can even contribute $6,500!
Starting in the year 2033 (presumably, that’s the calendar year when you turn 59.5) I assume you withdraw from your 401k.
In 2034, your consumption target falls because you have paid off the mortgage.
I solve for the withdrawals so the income net of taxes exactly matches the consumption target.

Results:

Despite the conservative assumptions you actually keep growing your taxable account by the time you reach age 59.5. No issue with running out of money in taxable accounts!
At age 70, when required minimum distributions become an issue you’ll have a pretty large 401k. Under today’s tax laws, this would likely push you into the 25% tax bracket because you’ll be forced to withdraw 1/27.4=3.65% of your 401k balance that year. As I have described in these case studies before, this is a good problem to have – too much money. And there’s not much you can do about it because you likely have very limited room for Roth conversions before age 70, see below.
But again: overall the picture looks extremely positive. Despite the withdrawals, and despite lowballing the expected returns, you will likely grow your capital even in inflation-adjusted terms!

Is Peer Street really worth it?

On the surface, it seems as though Peer Street beats equities. 8% return beats my conservative estimate of equities in the near term (5.75%) and even in the long-term (7%). But keep in mind that equities still enjoy a superior tax treatment. You can likely generate 7% returns in the long-term that will be taxed only at the state level, presumably around 8% marginal taxes, for a net return of 6.44%. However, your Peer Street returns of 8% will be taxed as ordinary income, presumably at 15% federal and 8% state. That 8% Peer Street yield is now only 6.16% after tax, less than equities.

If you are not too attached to your alternative investments you might consider scaling down your Peer Street engagement over time. I understand that the 8% yield offers a lot of peace of mind especially considering that equities seem so expensive at this point. But maybe if equities are a bit cheaper again in the future (measured by their CAPE ratio) you’ll revisit this option.

Another option would be to explore if Peer Street (or some other provider) offers IRAs. Then reinvest principal payments from Peer Street in an equity fund in the taxable account, but do the reverse transaction in your IRA (sell equity funds, invest in Real Estate loans) and thereby slowly switch the Peer Street investment to an IRA where you don’t have to tax the massive 8% yield as ordinary income every year.

Roth Conversions?

It sounds like your income combined with your husband’s will put right at the top of the 15% federal bracket. You may even slightly overshoot it. But there may be years when you stay below and then it would be crucial to max out the 15% bracket for Roth conversions. Remember, 15% is likely the lowest rate you ever pay for ordinary income on your federal return and at some point in the future, when required minimum distributions kick in you may pay 25% at the margin. So, if you haven’t done so already, check out the Case Study from a few weeks ago where I proposed how to implement Roth Conversions to max out the 15% federal bracket.

A word about Betterment

Your $277k combined investment in Betterment will cost you about $700 in annual fees. Betterment now charges 0.25% p.a.! And that’s on top of the ETF expense ratios! For that reason, I’m not a big fan of Betterment! I have written pretty scathing reviews before:

Shifting existing asset with built-in capital gains is a bad idea. Not applicable to your situation, but still a fun read: We just saved $42,000 by not switching to Betterment.
How to do Tax Loss Harvesting yourself without spending 0.25% p.a. on Betterment: Be Your Own DIY Zero-Cost Robo-Adviser!
How Robo advisers inflate their estimates of how much benefit you can generate from tax loss harvesting: Why we don’t use Robo-advisers.

Especially the third post is relevant to you. In the decumulation phase, the benefit of tax loss harvesting is hugely diminished. One could even argue that a few years into your retirement when you have stopped new investments and started living off your dividends, there will be no more tax losses left to be harvested.

But it gets even worse. Even under the most optimistic assumption that Betterment keeps generating $3,000 every year in tax losses they are much less useful in retirement. If you are already liquidating long-term gains to fund your expenses in retirement, note that your tax losses from Betterment will first be netted against taxable long-term gains before you can them to lower your ordinary income. In the worst case, this may be worth only $3,000 multiplied by the 8% marginal tax rate on long-term gains. You’d pay $700 in fees for $240 worth of tax savings.

Another reason to be cautious about Betterment/Wealthfront: They could be a tax audit time bomb! It’s hard to coordinate your Betterment portfolio with your other portfolio holdings outside Betterment. Thus, some of the tax losses generated by Betterment could be disallowed by the IRS. Betterment makes sure to avoid disallowed wash sales within the Betterment account. But if you bought sufficiently similar assets (e.g. VTSAX) in your other accounts (whether through new savings or dividend reinvestments) in the 30 days before or after Betterment harvested losses in the VTI index ETF, the IRS will disallow a portion or all of the tax losses.

My recommendation: Roll over the Betterment accounts to another broker, but do so in kind, i.e., without selling the underlying ETFs. Then in their new domicile (e.g. at Vanguard), you’d sell the bond ETFs (they likely have the lowest capital gains, too!) and shift to equity ETFs. It’s more tax-efficient to keep bonds in the tax-deferred accounts. Then start living off the dividend income and (potentially) sell some of the ETFs as needed to fund expenses.

Pay off the mortgage?

There was a good discussion in the ChooseFI Facebook group the other day. Applying average expected returns to calculate the pros and cons of prepaying the mortgage is misleading. Of course, an equity portfolio will, on average, beat 3.25%. But in safe withdrawal rate studies, we are concerned about the tail events when equities underperform. My preliminary research (to be published sometime in the future in the Safe Withdrawal Rate Series) shows that not having a mortgage helps with Sequence of Return Risk in retirement and will likely sustain a higher safe withdrawal rate. Again: you lose on average if you pay off the mortgage but in the worst case scenarios it’s priceless to be mortgage-free.

I think there is no need to sweat this issue because your finances have so much safety margin built-in that you probably don’t have to be too concerned about Sequence Risk. Also, unless you both agree to pay down the mortgage and you each pump $130,000 to pay down the mortgage, the matter is further complicated by the unusual arrangement with your husband. If you were to pay down your share ($130,000) your husband would have to pay off his remaining half but he’d have to make the full payment ($1,670) rather than his current $835 payment (though, for a shorter time, of course). But that may create a cash flow problem for your husband now. Maybe think about the mortgage payoff option once more when your husband retires in 7-10 years!

Conclusion

As you might have expected, your finances are in excellent shape. Well, they better be because you already pulled the plug and retired. When using the historical simulations, your current consumption needs stay below even the fail-safe withdrawal rate. It looks like you will sail (surf?) through early retirement with enough of a cushion to last for many decades and to leave a nice bequest to your daughter. I hope you find ways to use all that extra time productively. Maybe finally learn how to surf?! Best of luck to you and your family!