Our Finances

Our Finances

Our finances were a key consideration as we planned our nomadic life. They determined when we could quit work, how much we could pay for our motorhome and toad vehicle, where we could travel, as well as where we would stay and what we could do as we traveled.

Finances are not only very important, they can also be crazy complex. We are both experienced with Excel, and know how to run the numbers. Yet, we ended up spending countless hours crunching figures and scouring the internet for helpful articles.

Our goal was quite ambitious: optimize our Social Security benefits, minimize our income taxes, generate consistent additional income, extend the value of our investment portfolio to last our lifetime, and live within our budget.

We began our research years before retiring. Four authors were especially helpful: finance journalist at PBS, Paul Solman; economics professor at Boston University, Larry Kotlikoff; director of personal finance at Morningstar, Christine Benz; and certified financial planner, Michael Kitces. Their advice was particularly valuable in those areas where our finances intersected the government, such as Social Security and Income Taxes, which often defy common logic.

All in all, we’re satisfied with the way we’ve set up our finances. Of course, they are specific to our story, and certainly not perfect. But we’ve made them public with the hope that others nearing retirement or thinking of going nomad can also benefit from what we’ve learned and the decisions we’ve made, both good and bad.

This discussion of our finances involves four distinct factors, each one mathematically complicated in its own right. Note that all the calculations below use current figures, and are not adjusted for inflation.

• Living Expenses
• Social Security Benefits
• Additional Retirement Income
• Income Taxes

Living Expenses

Compared to the other three factors, estimating our living expenses was relatively easy. We’ve lived by a budget our entire marriage. We were an early adopter of Quicken, dating back to its beginning in 1983. So we simply had to transfer our past expenses to what we estimated they would be going forward.

For the most part, our expenses stayed roughly the same after retiring, with a few exceptions. There were big drops in income taxes, retirement savings, charitable giving, and auto expenses. These drops directly result from the reduction in our income following retirement, and would have happened whether we were stationary or nomadic.

Income Taxes: Before retirement, our federal marginal tax rate was 25% plus an additional 5% for the state of Ohio. We wanted to minimize our taxes in retirement. So we changed our domicile from Ohio to Florida, which has no state income tax.

We also reduced our total federal income tax rate to 0% by carefully controlling the amount of money we withdraw from our tax-deferred retirement accounts. This is by far the single largest drop in our expenses. For a more thorough discussion on how we managed to reduce our federal taxes to zero, see the detailed Income Tax section below.

Retirement savings: Before retirement, we were saving about 15% of our income for retirement. After we stopped working, the flow was reversed; we’re now taking money out of savings rather than putting it in. This change resulted in the second largest drop in our expenses.

Charitable giving: Before retirement, we gave away 10% of our gross income to charity, i.e. tithed. Now that we are retired, we no longer have our own income. Instead, we meet our living expenses with Social Security benefits and withdrawals from our retirement savings. Some of this money was tithed on previously, and some wasn’t.

For instance, Social Security is made up of equal contributions from employees and employers. Since we contributed the employee half from our gross income, we don’t tithe on that part again. But we do tithe on the employer half. The government uses a similar logic when taxing Social Security. See the detailed Income Tax section below.

Along the same lines, our retirement savings is made up of principle plus the interest and dividends we earned from that principle. Since we contributed the principle from our gross income, we don’t tithe on that again. But we do tithe on the income and dividends. Unfortunately, we haven’t kept good records on how much of our retirement savings we contributed and how much was earned. But for simplicity sake, we estimate it was half and half.

What this means in practice is that we tithe on half our Social Security income and on half of the withdrawals from our retirement savings. This change in how we tithe has resulted in the third largest drop in our expenses.

Auto expenses: Before going nomad, we had two cars. Now, we only have one, which we tow behind our motorhome. Changing from two cars to one saves us hundreds of dollars a month, and is the fourth largest reduction in our expenses.

We also added two new expenses directly related to our being nomads.

Campground expenses: The utility bills for our house, such as electric, gas, water, and sewer, used to run us about $250 a month. But as nomads, we no longer have to pay utilities since they are included in our campground fees. But we do pay for campgrounds, which runs about $600 a month.

Motorhome expenses: Our house mortgage was higher than our loan payment on our motorhome by $200, but our maintenance and repair costs are higher on our motorhome than on our house by $400.

The bottom line is that it’s costing us $550 a month more to live as nomads than if we had stayed stationary: $350 for campgrounds and $200 for the motorhome. But overall, the reductions in expenses from being retired far outweigh the increases from being nomadic. Our budget is now $4,400 a month, which is less than half of what it was before retiring.

Social Security Benefits

Social Security is one of those subjects that everyone knows about but almost no one understands, like gravity or the internet. When we first began looking into it, we just wanted to know when and how to file for Social Security. This quickly led to an investigation of why to file when and how.

It turns out that nearly everyone files when they retire, which seems to make sense. That was our first option. But we learned we could get a lot more.

Benefits are maximized when retirees wait until age 70, assuming they live long enough. The problem is that when we retired, Becky was 62 and I was 65. That meant I would have to wait for five years and Becky for eight. Maximizing was not a viable option. We simply did not have the means to wait eight years to start her benefits.

So we chose a middle way: Becky took her benefits at retirement; I’ll wait until age 70. A little now and a lot more later. Instead of maximizing, we’re optimizing.

Becky’s retirement benefit is $1,460 a month. The advantage of her filing at 62 is that we started getting benefits as soon as we retired. The advantage of my waiting is that my benefits are increased by 41% from what they would be if I had taken them when I retired.

Here’s the breakdown: if we both took our own benefits when we retired, our combined amount would be $3,210 ($1,460 + $1,750). By waiting until I’m 70, they’ll be $3,920 ($1,460 + $2,460), which is a difference of $710 a month.

What tipped the scales for us is that I was allowed to file a restricted application for a spousal benefit of $952 a month based on Becky’s earnings. My spousal benefits started at my age 66. It’s like getting $45,696 extra just for waiting. This gave us a combined benefit amount of $2,412 ($1,460 + $952) a month. The net result is that over 25 years, we’ll get $1,104,624 from Social Security for my waiting versus $963,000 for not waiting. That’s an increase of $141,624.

Note that because of a rule change in 2015, filing a restricted application for a spousal benefit is no longer available to people born in or after 1954. That excludes Becky. If she could have gotten a restricted spousal benefit, it would have been an enticing option for her to wait until age 70, and for me to begin at retirement. Just a reminder that Congress can change the rules whenever.

As a general guideline, when the two spouses are about the same age, the one with the higher earnings should wait to age 70. Otherwise, the older one should wait, especially if the older spouse qualifies for a restricted spousal benefit. But each couple is unique, and their strategy should be tailored to their particular situation.

The other big advantage of my waiting is that one retirement benefit is much larger than the other, $1,460 for Becky vs $2,460 for me. Not only does this give us more money while we both live, it also provides more when one of us dies because the survivor keeps the larger amount as his/her survivor benefit. In other words, if I had taken my retirement benefit when I retired, the survivor benefit would have been $1,750. By waiting, it could be as high as $2,460 depending on my age at death.

We’ve seen many examples among our family and friends where one spouse outlived the other by a decade or more, and we know firsthand how important the survivor benefit is. When a spouse dies, their retirement check stops, too. But in our case, at least the survivor could have an extra $710 a month because I waited.

Of course, waiting has one huge disadvantage: namely we have to draw out more from our retirement savings for living expenses from now until I’m age 70. Specifically, it’s costing us a total of $37,300 to wait.

We see the cost of waiting as buying a low cost, inflation adjusted, deferred “government annuity” that pays $710 a month for life starting at my age 70. For comparison, it would cost us $125,000 to buy a similar Joint Life Annuity from an insurance company with the same payout.

Like any annuity, if we both die before the breakeven point, we won’t get our money’s worth. But if either of us lives longer than that, the annuity will have proven a good choice. That’s especially true if we both live longer than the breakeven point, which for us (because I’m getting the spousal benefit) is less than ten years away.

Three out of four of our parents lived to advanced ages, so living long is a real possibility for us. Take a look at this joint life expectancy chart based on our current ages taken from Vanguard’s online calculator.

Probability of Either of Us Living X Additional Years
Additional Years Probability
10 years 97%
15 years 90%
20 years 75%
25 years 49%
30 years 23%

Even without taking heredity into account, there is a 97% likelihood one of us will hit the breakeven point of 10 more years. There’s a 49% likelihood that one of us will live another 25 years. And a 23% likelihood that one us will live another 30 years. There’s even a 1% chance I’ll live to age 100 and a 4% chance Becky will. That’s why we’re buying our “government annuity”. Not because we will live long, but because we might.

Once I turn age 70, the gap between our income and expenses narrows dramatically, as my benefits switch from a small spousal benefit to a much larger delayed retirement benefit. The amounts drained from our retirement savings will drop from a couple thousand dollars a month to a couple hundred. At that time, the dividends and interest from our remaining retirement investments should be enough to make up the difference.

Over the next 25 years, we expect to spend $1,320,000 ($4,400 a month for 300 months). Amazing isn’t it! Thankfully, Social Security is going to provide most of that. If we had both taken our benefits at retirement, we would need to draw out $357,000 from our savings to cover the difference. But by waiting, we’ll only need $215,000. Put another way, the cost of delaying is $37,300, while the potential value is $142,000 extra during our lifetimes (assuming we both live for another 25 years).

But what if we don’t live another 25 years. Well, if either of us dies before I reach age 70 (which financially is the worst case scenario), the picture will not be as good as outlined above. But waiting is still better than not waiting, if only for the larger survivor benefit, which grows by 8% for each year I wait.

Not surprisingly, deciding when to take our survivor benefit is as complicated as everything else about Social Security. The amount and timing of the benefit depend on two factors: the age of the deceased and the age of the recipient. There are three possible scenarios.

(1) I die before Becky reaches age 66. I am almost three years older than Becky, and it is certainly possible that I might die before she reaches her full retirement age. The amount of her survivor benefit will be based on whatever my retirement benefit would be at the time of my death, which will be more than $1,860, since I would be over age 66, but less than $2,460, since I would be under age 70.

For example, if I died when Becky was age 65, I would be age 68. My retirement benefit would be about $2,160, and Becky would be entitled to that amount as her survivor benefit.

However, Social Security reduces the benefit by 4.7% for each year it’s taken before the recipient reaches full retirement age. In the example above, if Becky took it immediately at age 65, it would be reduced $2,045 for taking it one year “early”. And the reduction would permanent.

Therefore, she should delay her survivor benefit and continue taking her current retirement benefit of $1,460 until she reached her full retirement age of 66. Then she should switch to her full survivor benefit of $2,160.

(2) Becky dies before I reach age 70. On the other hand, Becky might die before I have started taking my delayed retirement benefit. If so, my survivor benefit will be based on her current retirement benefit of $1,460.

Fortunately, since I am already over age 66, the amount won’t be reduced. Therefore, I should give up my $952 spousal benefit and immediately start taking my survivor benefit of $1,460.

Then when I reach age 70, I should give up my survivor benefit and switch permanently to my own retirement benefit as planned, which will have grown to $2,460.

(3) Neither of us dies before I’m age 70. For obvious reasons, this is the most hoped for outcome. But eventually one of us will die. Regardless of who survives, he or she will receive $2,460. This is because Becky will be past age 66 and I will be past age 70. As mentioned earlier, getting the larger survivor benefit is the main advantage of my waiting to take my retirement benefit until age 70.

Just a word about the solvency of Social Security. The government estimates that the Social Security Trust Fund will be solvent until 2034. I’ll be 83 and Becky will be 80. So personally, we’re not too worried about it. Furthermore, Congress passed a law a while back that says current retirees cannot have their benefits cut. If the Trust Fund doesn’t have the money, the shortfall will come from the Treasury. Hopefully, Congress will fix the problem before it gets that far. It would be a financial disaster for many younger people if they don’t get what is due them.

Additional Retirement Income

Our strategy for taking Social Security outlined above shows how dependent we are on having additional retirement income, especially in the early years. For most of our marriage, we saved between 5% and 10% of our income for retirement. As we approached retirement, we increased that amount to 15%.

When we were young, our asset allocation was heavily weighted toward large cap stocks, mostly in S&P 500 index funds. As we got older, we diversified more and shifted some assets toward dividend paying stocks and interest bearing bonds. By the time we retired, our asset allocation was 60% stock funds and 40% bond funds.

One of the legacies of working for various employers over decades is that we had multiple retirement accounts spread across multiple financial institutions, including a 403(b) at Guidestone, two 401(k)s at Fidelity and one at Vanguard, a SEP at Schwab, and two Rollover IRAs and two Roth IRAs at Vanguard. Within these accounts, we had a dozen stock funds, some REITs, and several bond funds.

After we retired, the first order of business was to consolidate all these retirement accounts at Vanguard. We chose Vanguard because of their wide investment selection and low management fees. We’ve been fans of them for many years. Having all our investments in one place simplifies asset management greatly. Each of us now has just two accounts: a traditional tax-deferred IRA and a tax free Roth IRA, four accounts total.

We moved some of the bond funds to long-term cash, specifically the Vanguard Short-Term Inflation-Protected Securities Fund (a.k.a. TIPS). This fund is kept in a traditional tax-deferred IRA account. Our goal is to keep enough long-term cash available to meet at least 36 months of net living expenses, i.e. our monthly expense budget minus our Social Security income.

We also have a savings account with Ally Bank to keep enough short-term cash to meet 12 months of net living expenses. We like Ally because it’s very easy to use on the road, and their savings account pays one of the highest interest rates in the county with no minimum.

We replenish the cash accounts with income from our investments as it becomes available. Specifically, each time our investment portfolio increases by $10,000, we will move half the gains to cash and let the other half remain invested. This allows us to replenish our cash accounts while still letting the portfolio grow naturally without adding new money.

Having a four year supply of cash should allow us to weather a major downturn in the market without being forced to sell any of our investments at a loss. In other words, our cash should only come from gains, never principle, unless we are desperate and completely out of cash.

In addition to these cash accounts, we have some gold and silver coins, along with a bit of cryptocurrency. These can also be easily sold to meet net living expenses if necessary without hurting our investments.

Update: In late 2017, we converted some of our gold and silver coins into additional cryptocurrency, and sold the rest for cash. Overall, our crypto investments produced significant short-term capital gains, which we were required to report as taxable income even though they were mostly on paper. In the future, we plan to hold these investments for more than one year so they can be claimed as long-term capital gains in order to take advantage of the lower tax rate.

Besides the cash accounts, our investment portfolio is now made up of just three balanced Vanguard mutual funds (two actively managed and one index fund) compiled from the jumble of stock and bond funds we used to have. Some of the portfolio is held in our traditional IRAs and some in our Roth IRAs.

70% of our portfolio is in these two Vanguard active funds:
Wellington Fund (65% stocks / 35% bonds)
Wellesley Income Fund
(35% stocks / 65% bonds)

And the other 30% is in one of these Vanguard index funds (which one depends on market conditions):
LifeStrategy Growth Index Fund (80% stocks / 20% bonds)
LifeStrategy Moderate Growth Index Fund
(60% stocks / 40% bonds)
LifeStrategy Conservative Growth Index Fund (40% stocks / 60% bonds)
LifeStrategy Income Index Fund (20% stocks / 80% bonds)

Wellington and Wellesley are near mirror images of each other, which is hinted at in their fund names. (Arthur Wellesley was the 1st Duke of Wellington, who led the defeat of Napoleon at the Battle of Waterloo.) The two funds are run by the same team of portfolio managers and we treat the pair as a single investment, which we call the “Wells.”

Wellington allocates 65% to stocks and 35% to bonds, while Wellesley is 35% stocks and 65% bonds. The bond portions are the same except for the weightings, but the stock portions are from mostly different companies. Wellington holds 100 large cap value stocks, while Wellesley holds 60 large cap dividend stocks. About 20 stocks are in both.

The “Wells” Stocks Bonds
Wellington
(60 stocks / 900 bonds)
65% 35%
Wellesley Income
(100 stocks / 900 bonds)
35% 65%
Combined
(140 stocks / 900 bonds)
50% 50%

Though investors usually buy one or the other, we have equal amounts of each. This results in a combined allocation of 50% stocks and 50% bonds, and forms the core asset of our portfolio.

The “Wells” have very long track records. Wellington was established back in 1929, and was in fact the first balanced mutual fund. Wellesley began in 1970 specifically to complement Wellington. Together, these two funds have $152.5 billion in assets. And though they are mirror opposites, they tend to move in the same direction, which is usually up. Unlike index funds, these actively manged funds enable the managers to move some assets out of stocks into bonds or vice versa depending on market conditions.

The bond portion of both funds use the Bloomberg Barclays U.S. Credit A or Better Bond Index as a benchmark, which is composed of high quality bonds, mostly investment grade corporate issues that generates income and protects against stock market downturns.

The funds can deviate from their bond allocations by 5% up or down. For instances, Wellington wisely dropped it’s bond allocation, which is normally around 35%, to 31% in 2016 as interest rates started rising. Wellesley reduced its bond holdings a bit as well and shifted to a larger share of high yield bonds. I would expect that once higher interest rates have stabilized, both of these funds will return to their usual allocations.

The stock portions of each fund use different benchmarks. For Wellington, it’s the S&P 500 Index. For Wellesley, it’s the FTSE High Dividend Yield Index, which is the top yielding half of all dividend paying stocks.

The different benchmarks reflect different goals, but there is some overlap. Wellington’s is to “provide long-term capital appreciation and reasonable current income.” The fund has about 100 stocks of what they consider to be the cream of the crop from the S&P 500 in terms of growth and income.

Wellesley’s goal is to “provide long-term growth of income and a high and sustainable level of current income, along with moderate long-term capital appreciation.” The fund has just 60 stocks that have historically paid a larger-than-average dividend or that have expectations of increasing dividends. In other words, the “Wells” are very blue chip, with Wellington being slightly more tilted toward stock appreciation.

Despite being fans of index funds for a long time, we chose the “Wells” as the core of our portfolio precisely because they are not index funds. Being actively managed allows the managers to have leeway in terms of both which individual stocks and bonds to include as well as how much to allocate to each type of asset while still maintaining a conservative investment approach. They have produced positive results in both good times and bad.

We like this conservative, yet flexible, approach so much that we seriously considered holding the “Wells” as our sole non-cash investment. However, the “Wells” only cover a rather narrow part of the market. We like the emphasis on large U.S. companies, but wanted some exposure to small and mid cap stocks as well as some additional international exposure. So in the end, we decided to add that diversification with one of the LifeStrategy Index Funds.

There are four LifeStrategy Index funds to choose from: Growth (VASGX), Moderate Growth (VSMGX), Conservative Growth (VSCGX), and Income (VASIX). Each LifeStrategy fund is actually a fund of funds made up from the same four underlying Vanguard “Total” index funds, but with different weightings. They are very broadly diversified, essentially covering all the publicly traded stocks and bonds in the world. Can’t get broader than that.

Below is a chart showing how the LifeStrategy funds are allocated among the four Vanguard “Total” index funds.

LifeStrategy
(Fund of Funds)
VASGX
80/20
VSMGX
60/40
VSCGX
40/60
VASIX
20/80
Total Stock Market Index
(3,600 U.S. stocks)
48% 36% 24% 12%
Total International Stock Index
(6,200 foreign stocks)
32% 24% 16% 8%
Total Bond Market II Index
(8,600 U.S. bonds)
14% 28% 42% 56%
Total International Bond Index
(4,400 foreign bonds)
6% 12% 18% 24%

Currently our asset allocation is 30% in LifeStrategy Growth, 35% in Wellington, and 35% in Wellesley. This results in an overall stock allocation of 60%. When combined with our cash reserves, our total portfolio’s stock allocation is closer to 50%, with which we are quite comfortable. We are willing to shift from Growth to one of the other LifeStrategy funds with more bonds if market conditions warrant or our risk profile changes.

Because of their worldwide diversification and tight adherence to the underlying indexes, the LifeStrategy funds are quite stable. The underlying funds are common choices in many 401k plans. That means even in a down market, the funds will still have more inflows than outflows, which keeps their values from falling too quickly in a bad market.

The underlying funds have an astonishing $1.1 trillion in assets. A lot of people, including many retirees, put their trust in these funds. It doesn’t mean they are right, but it is comforting to know that we are not alone. Plus the incredible size itself adds stability. It takes a long time to unwind a trillion dollars.

The bottom line is that since the LifeStrategy funds tend to move rather slowly, we should have ample time to make any adjustments without selling at a loss. Our four year cash cushion means that even if we find ourselves on the wrong side of the market, we should have enough cash to weather the storm.

Given all this, we think LifeStrategy is a perfect complement to Wellington and Wellesley. We believe that this combination brings together Vanguard’s best actively managed funds with their best index funds in a safe, balanced, and easily adjustable split between stocks and bonds. It provide us with exposure to assets worldwide while still focusing on the largest and most profitable U.S. corporations, all with a low average expense ratio of 0.14%.

Update: Vanguard began offering global versions of Wellington and Wellesley in November 2017. We may decide to shift some of our “Wells” money to the global versions as we learn more about them. In the meantime, we are continuing to keep a significant portion of our portfolio in LifeStrategy in part because of its international exposure.

What we like most about this portfolio is whether the markets are rising, going sideways, or falling, these funds will produce income, even if the portfolio as a whole isn’t appreciating. While there is no such thing a a bullet proof portfolio, we believe that this comes pretty close.

We understand that this portfolio will miss out on the big fast swings up, but it should also miss the big fast swings down. And that’s fine with us. Our guiding principle during retirement is “as long as we don’t lose much money, we’ll should have time to recover, and hopefully won’t run out of it.”

We believe our portfolio will provide us with sufficient control to adjust to changing conditions, while producing enough income to meet our needs going forward without seriously decreasing the principle or incurring any more risk than necessary.

Since each fund is a balanced fund in its own right, most of the rebalancing between stocks and bonds will be done automatically by the fund managers. If any of the funds grows out of balance, we can bring it back in balance manually each time we cash some out. 

In addition, periodically throughout each year, we will convert some of our tax-deferred accounts to our Roth accounts using the same fund allocations or take a distribution in cash, but only when we can do so without paying income taxes. The section below explains how we determine in advance the amount we can withdraw from our tax-deferred accounts each year without paying any income taxes.

Income Taxes

As mentioned above, our Social Security is not enough to cover all of our living expenses. So each month we have to withdraw some funds from our retirement accounts as discussed above.

Most of our savings is in tax-deferred accounts, which means that any withdrawals are included in taxable income. But as long as our total taxable income is less than our tax deductions, we can make the withdrawals tax free.

With a bit of planning, converting tax-deferred assets to tax free is quite possible. But it’s not simple, mostly because taxable income causes some Social Security income to become taxable as well. In other words, the more you withdrawal from your tax-deferred account, the more you pay in taxes, not just on the withdrawals but also on Social Security, unless you stay within certain limits.

Let’s start with some definitions:

Ordinary income: wages and salaries, taxable interest and dividends, taxable pensions and annuities, capital gains distributions, and distributions from a tax-deferred retirement account. All ordinary income is taxable income.
Provisional income: a measure used by the IRS to determine the amount of Social Security that is taxable. It includes ordinary income, half of Social Security, and tax-exempt interest from municipal bonds.
Taxable Social Security: an amount based on how much provisional income is over certain thresholds set by Congress.
Taxable income: the sum of ordinary income plus taxable Social Security.
Personal Exemptions: As of tax year 2016, each person is entitled to a personal tax exemption of $4,050.
Standard Deductions: As of tax year 2016, standard deductions for a married couple filing jointly is $12,600. Persons over age 62 receive an additional $1,250 each. (See chart below.)
Taxed Income: all taxable income, minus deductions and exemptions. Not all taxable income is taxed.

We initially assumed that income from Social Security was 100% tax free, and that 100% of the withdrawals from our tax-deferred retirement accounts would be taxed. However, we were only partly right on both counts.

Yes, it’s true that Social Security is usually tax free. But adding in ordinary income can cause up to 85% of Social Security to become taxable. And it’s true that withdrawals from a tax-deferred retirement account are taxable, since they are treated as part of ordinary income. But if taxable income is less than or equal to deductions and exemptions, none of it is taxed.

Our total deductions and exemptions are $23,200. That means we can have up to $23,200 of taxable income without owing taxes. Below are the 2016 standard deductions for each filing status.

Filing Status Standard Deductions
Single $6,300
Married Filing Jointly $12,600
Married Filing Separately $6,300
Head of Household $9,300
Qualifying Widow(er) $12,600

Update: In 2017, Congress eliminated the personal exemptions and raised the standard deductions to $12,000 for singles and $24,000 for couples.

Armed with this new information, we decided to see if we could keep our ordinary income from all sources high enough to live on but low enough to owe no taxes.

The first challenge is to determine how much, if any, Social Security is taxable. As mentioned earlier, Social Security is made up of equal contributions from employees and employers. Our half was taxed back when the income was earned, but the employer half was not. Therefore, the government feels it can legitimately tax half our Social Security benefits. Fair enough. But they have a strange way of doing it.

The IRS uses provisional income to determine what amount of Social Security is taxable. If it exceeds the lower threshold set by the IRS, a complicated formula is applied to determine how much of Social Security is taxable.

Below is a chart showing provisional income thresholds for each filing status and the corresponding percent of taxable Social Security, followed by an explanation of the formula for determining taxable Social Security.

Filing Status Provisional Income Thresholds Social Security Taxation Percents
Single Filers Less than $25,000 0%
$25,000 – $34,000 Up to 50%
More than $34,000 Up to 85%
Joint Filers Less than $32,000 0%
$32,000 – $44,000 Up to 50%
More than $44,000 Up to 85%
Separate Filers $0 Up to 85%

Using the provisional income thresholds and taxation percents from above, the formula for calculating taxable Social Security for joint filers like us is as follows:

Social Security Taxation Formula
Step 1 Determine the amount of provisional income — half of Social Security benefits plus all ordinary income plus tax-exempt interest from municipal bonds. Income and withdrawals from Roth IRAs are not included.
Step 2 Determine the amount of provisional income between the lower threshold of $32,000 and the upper threshold of $44,000. Multiply the result by 50%.
Step 3 Determine the amount of provisional income above the upper threshold of $44,000. Multiply the result by 85%.
Step 4 Determine the minimum taxable Social Security by adding the results of Step 2 and Step 3.
Step 5 Determine the maximum taxable Social Security by multiplying your total Social Security benefits by 85%.
Step 6 Your taxable Social Security is the lesser of the results in Step 4 or Step 5.

This convoluted method of taxing Social Security is often referred to as a tax torpedo. It catches many retirees who have non-Social Security income by surprise and blows their financial plans sky high. A retiree with a small amount of ordinary income normally taxed at 10% or 15% can drag in an extra 50% of Social Security, effectively raising their tax rate to 15% or 22.5%. And a higher amount normally taxed at 25% can drag in an extra 85% of Social Security, raising their effective tax rate to 46.25%. It’s crazy.

The tax torpedo is especially dangerous for retirees who leave too much of their savings in a tax-deferred retirement account. This is because of Required Minimum Distributions (RMDs). By law, retirees must take RMDs from their tax-deferred retirement accounts starting at around 4% a year at age 70½ and increasing each year thereafter. Essentially, the government forces retirees to cash out tax-deferred accounts so they can collect the taxes. But the result is that a large chunk of Social Security is taxed, too. Kaboom!

Congress should be ashamed for taxing Social Security this way. I think it would be much fairer if retirees simply included the employer half of their Social Security benefits as ordinary income. Essentially, it’s the same as treating the employer half like a tax-deferred IRA and the employee half like a tax free Roth IRA. Most retirees would still pay no income tax, because the employer half of their Social Security even with small amounts of other income, would be less than their deductions and exemptions. Wealthier people would pay their fair share. And no one would get blown up by the tax torpedo.

We plan to avoid a tax torpedo by keeping our taxable income equal to our deductions and exemptions. The idea is to slowly transform our taxable income into tax free income a little each year.

Let’s review the basics. If we had no additional income, including no Social Security, we could withdrawal $23,200 from our tax-deferred retirement accounts without owing any taxes. Of course, we do have Social Security. So the withdrawals cause some of our benefits to be included in taxable income. This in turn requires us to lower our withdrawals if we want to owe no taxes.

As defined earlier, taxable Social Security is based on provisional income, which equals ordinary income plus half of Social Security. And taxable income equals ordinary income plus taxable Social Security. As you may suspect, the equation is circular. The problem is that when ordinary income goes up, taxable Social Security goes up, which is based in part on ordinary income.

If that last paragraph gave you a headache, you have good reason. When confronted with a tax system like this, it’s hard to tell whether the government is stupid or evil. Perhaps we should demand that government officials do their own taxes.

Unfortunately, there’s no way to know beforehand exactly how large your ordinary income can be, specifically the withdrawals, and still avoid owing taxes, without doing advanced calculations. However, if you are a joint filer and just want a rough estimate, simply subtract half your Social Security and tax-exempt interest from the lower threshold. The result is tax free ordinary income, assuming it’s less than your deductions and exemptions.

The weakness of this method is that it doesn’t take full advantage of all your deductions and exemptions, and can miss the mark by thousands of dollars. If you want to know the maximum amount of ordinary income you can get tax free, you’ll likely need to use a tool like Microsoft Excel.


WARNING: The next few paragraphs are quite technical. Those unfamiliar with advanced features in Excel may want to skip ahead.


Using the Solver Add-in for Excel, we can figure out the maximum amount of ordinary income we can have in addition to our Social Security without owing taxes. No guesses, no surprises, no torpedoes.

If you’re not familiar, Solver is part of the “what-if analysis” suite of free add-ins for Excel. It uses an iterative process to find the value for one cell limited by constraints set by values or formulas in other cells. It’s a bit of magic.

First, we begin by setting our initial variables. Three are known: Social Security, the maximum taxable Social Security (which is 85% of it), and deductions and exemptions, which in our case is $23,200. One variable is unknown: ordinary income, which we will initially set to zero. This is what Solver will calculate for us.

Second, we list our constants: the two provisional income thresholds ($32,000 and $44,000) and the two taxation percents (50% and 85%).

Then, we create a series of formulas. The first calculates our provisional income. A second calculates the taxable Social Security based on the lower threshold. A third calculates the taxable Social Security based on the upper threshold. A fourth adds the results of the lower and upper calculations to determine the minimum taxable Social Security. A fifth determines the actual taxable Social Security by choosing either the maximum or the minimum amount, whichever is less. And a sixth adds ordinary income with taxable Social Security to determine our taxable income. A change in any of the initial variables automatically changes the results of every formula.

The final step is to activate Solver. It changes the value of the cell containing ordinary income to the maximum amount it can be so that our taxable income is exactly constrained by, i.e. equal to, our deductions and exemptions.

As a follow up, we will subtract any ordinary income received from other sources, such as wages, interest, etc., from the maximum amount of ordinary income Solver has calculated. The remaining amount is what we can withdraw from our tax-deferred retirement account for the year without owing any taxes.


This ends the technical section.


Below are the actual numbers I’ve calculated for us using Excel and painstakingly double checked with the latest IRS worksheet, Publication 915 for tax year 2016. You’re welcome to check it out yourself.

The goal is to have exactly $23,200 in taxable income that is offset by $23,200 in deductions and exemptions. Currently, we receive $29,196 in Social Security benefits. If we added $23,200 in ordinary income, it would trigger $2,899 in taxable Social Security, and we would owe taxes.

Instead, if we reduce our ordinary income to $21,267, only $1,933 in taxable Social Security is triggered. Our taxable income now equals $23,200 ($21,267 + $1,933). The result is that our total tax free income is $50,463 ($29,196 + $21,267).

Our annual budget is $52,800 ($4,400 a month x 12 months). That means we are $2,337 short ($52,800 – $50,463). Fortunately we have sufficient tax-free savings to cover the shortfall.

After I turn age 70, our Social Security benefits increase dramatically to $47,040. So we’ll need to drop our ordinary income to $18,293, which triggers $4,907 in taxable Social Security. But once again, our taxable income will equal $23,200 ($18,293 + $4,907), which means that our total tax free income will be $65,333 ($47,040 + $18,293).

At that point, we will have more than we need for our budget. Ironically, at the same time, the IRS will require us to begin RMDs. They don’t allow Roth conversions to be made with RMDs. So the RMDs will simply be withdrawn and deposited in a standard savings account. But any amounts we wish to withdraw above the RMD can be converted to a Roth, where it will live happily ever after, compounding tax free until we need it or we die.

Because we will be maxing out our Roth conversions each year, our actual RMD amounts will get smaller in dollars, even though they will get bigger as a percentage of the decreasing unconverted balance. Put simply, no torpedo.

If we have expenses that exceed our available tax free cash assets, we could withdraw more from our tax-deferred accounts. But it would have to be enough to cover the expenses as well as the increased taxes. For instance, if we need an extra $10,000, we’ll have to withdraw $11,765 ($10,000 for us and $1,765 for Uncle Sam). That’s a tax rate of 17.65%. It might be cheaper to borrow the money, and pay it back with future tax free withdrawals.

By making partial Roth conversions each year, all our tax-deferred savings will eventually be transformed into tax free savings. One nice feature about Roth conversions is that if we convert too much, we can put some of it back. This is called recharacterization, and can be done even after filing our taxes. However, recharacterization may be done away with by Congress at some point in the future.

I should point out that the figures above do not account for inflation. The assumption is that as our budget grows with inflation, so will Social Security and our standard deductions and exemptions.

Keeping everything in balance is not easy. One of our Social Security checks comes on the middle of the month, the other at the end. Income from our investment portfolio comes quarterly. But our expenses recur monthly. And taxes are done each year. Just keeping the timing straight is staggering.

Adjusting our withdrawals is particular challenging because when Social Security increases, larger amounts of it become taxable. This in turn lowers the amount of retirement savings that can be withdrawn tax free.

The chart below shows the interaction between increased Social Security and decreased additional tax free income. Figures assume combined benefits for a married couple.

To clarify, taxable income, which is taxable Social Security plus ordinary income, does not cause any taxes to be owed as long as the taxable income is less than deductions and exemptions.

Summary

Living expenses, Social Security, retirement savings, and income taxes. Each piece affects the other. In order to have a secure retirement, we need to manage them wisely. The stakes are high. A slip-up at any point could be ruinous.

Our budget is set. Our Social Security is set. Our initial asset allocation is set. Our tax strategy is set. These choices have been made. All that remains is managing the ongoing reallocation of our retirement savings.

Every month, every quarter, and every year for years to come, we’ll be reallocating these retirement assets — harvesting some of the income and growth from our Balanced Portfolio into cash and converting our Tax-Deferred assets to Tax Free ones.

Below is a simple chart that visually guides us in managing these asset flows. Everything starts in the upper left and slowly moves down or over. Our investment gains move from left to right to meet living expenses. Our tax free conversions move from top to bottom throughout the year to take full advantage of tax deductions.

Our Asset Flows
Traditional IRA
Balanced Portfolio
Tax-Deferred
Traditional IRA
Long-Term Cash Assets
Tax-Deferred
 
Roth IRA
Balance Portfolio
Tax Free
Savings Account
Short-Term Cash Assets
Tax Free

We believe that setting up our finances as outlined above has allowed us to optimize our Social Security benefits and minimize our income taxes, while generating consistent additional income and extending the value of our investment portfolio to last our lifetime. Wish us luck. And good luck to you.