This is the time of year when many are thinking about their annual tax returns, however, anytime is a good time to stress the effect of taxes on your spendable retirement income. If you have all your retirement income in municipal bonds and Roth accounts, which aren't taxable, then you will probably find nothing of interest here. Also, if you have a complex taxable account with a sizable portion in MLPs, REITS, and other non-qualified dividend payers, then you should realize that your tax requirements may be more or less than the norm.
I will stress up front that I am not a CPA or tax accountant so this should not be considered tax advice. I do however volunteer each tax season for an IRS sponsored group called VITA, which stands for Volunteer Income Tax Assistance. In that capacity I either complete or review hundreds of returns each tax season and this will be my fourth year volunteering for this group. All that is to say, I do have some knowledge in this field and especially as it relates to the retiree. One of the reasons I mention the retiree is that a lot of the returns that I see are from retirees.
In this article I will study not only the general effect of taxes on retirement but also how they can be forced on the IRA owner, whether you need income from the IRA or not. I use the phrase "can be forced on the IRA owner" above very specifically because after age 70 you will need to take the IRS required minimum distribution from your IRA account as I explain below. For now let's review the rules for the required minimum distributions from IRAs, or RMDs as they are frequently called. In a traditional IRA, 401k, or other tax-deferred account the IRS allows you to put money into the account tax-free, but insists that you start taking the money out at age 70 ½ or suffer a 50% penalty if you don't. The withdrawal rates in the most frequently used table start low at age 70 with a rate of 3.65%, but by age 90 that rate has risen to 8.8%. The good news is this is the exact withdrawal rate that I used in what I called my RMD strategy in a previous article, so we can now look at how that strategy, which attempts to maximize your retirement income, may be affected by taxes. First we need to establish some rules for what we might expect in the way of increasing deductions, exemptions, and expanding tax brackets in future years. I realize there is literally no chance that I can predict future tax rates any more than I can predict future stock prices, however, maybe there is a chance I will be lucky. (It was recently proven to me that luck actually does exist when I recently won first place in picking the 2014 year-end S&P500 price in a field of over 500 entries.) However, I digress; we are here to establish rules for tax rates over a 30-year timeframe, essentially from age 70 to 99 during a person's retirement; to be more specific, a married couple filing a joint tax return. If a single was to have the same income, in most cases their taxes will be higher, except in the case where their taxes are actually zero. The only way that luck plays a factor in this is in the future of tax law changes.
Since I like to at least extract theories from facts, let's look at the facts over the last five years as it relates to four important tax variables. Below is a table of some important tax rates, exemptions, and standard deductions for the years from 2011 to 2015:
From the above data I have decided to use the inflation factor of 2% for the personal exemption, standard deduction and the top of each of the tax rate brackets. In this way there is a better chance to at least simulate some inflation in the way the taxes are calculated. I realize that this can all be struck down with new tax legislation, but we need to start somewhere. In the table above I also make comment on the tax levels on long-term capital gains, as a reference for those with dividends or long-term capital gains in a taxable account. Note, I did leave off one LTCG tax bracket of 20% for those in the 39.6% tax bracket, mainly because there is no income in that range in this study. Also, the significance of the parameter called Social Security Base Amount (in this case $32,000) is that this is the amount for which if you take one-half of your Social Security income and add it to your other taxable income, when the total of this is greater than the $32,000 base, your Social Security income will start being taxed at a rate that moves from zero to 85%. The result is that once you hit this threshold of taxable Social Security your taxes are increasing at a higher rate than would be expected until you reach the 85% ceiling of taxes on your Social Security.
Once again I will use Johnson & Johnson (NYSE:JNJ) for this article. If you remember in my last article I gave results for three different start dates (1973, 1975, & 1980) covering 35 years each. In this article I will use the median result, which was the 1975 start date to study the effect of taxes on both the income-only investor and the second investor following the RMD strategy; however, in this case I will start five years into the previous study at age 70. I will use the current tax rates for 2014 to start the test at age 70 for the investor, which would correspond to the year 1980 in my table of results for starting year 1975.
The purpose of the study is to not only see how much tax the investor using the RMD strategy would have to pay, but also if it is even feasible for the income-only investor to stay with an income-only strategy, or will it be necessary to sell some investments to pay the tax bill. To do this, two investor accounts will be studied; one wishing to live from the dividends only and the second using a total return RMD strategy to maximize the withdrawals during retirement. The ground rules for the RMD investor are the same as in my previous article, except that whatever is withdrawn from the account needs to pay the taxes first and then the rest is income to the investor. For the income-only investor, the ground rules are to withdraw the RMD amount, pay the taxes as necessary from the dividends and the RMD if necessary, spend any dividend income left over and invest the rest back in JNJ stock in a taxable account, to continue to receive as much dividend income as possible. For those thinking why not just transfer the unneeded shares directly to the taxable account; you could do that, but it essentially complicates not only the investor's paperwork, but also my math in the study, so I opt for the cleaner solution. Do the withdrawals 4 times a year on the dividend pay date, just like in my previous article.
- In year 1 of the 30-year study both investors will be 70 years old in May and thus need to take an RMD.
- Both accounts start with $244,692, which is the previous year-end value for the income-only strategy from my previous work. This will not penalize the income-only investor for having a higher balance right out of the gate and simplify the math, because if their totals are the same to start, they will be forced to withdraw exactly the same amount each year from the IRS RMD rules.
- Both cases calculate the coming years RMD need by dividing the IRS RMD table divisor into the previous December 31st total account value. This is the same as my previous article, except now taxes must be paid.
- Both cases now take one-fourth of the above calculated RMD yearly withdrawal on each dividend pay date by first using the dividends and then selling shares as necessary to pay the tax.
- Real historical price and dividend data is used for JNJ, and the data comes from both Yahoo Finance and the JNJ website.
- For the purposes of the JNJ price and dividend data, this study starts at the end of 1979 with a value of $244,692 in both accounts when both taxpayers are 69 years old.
- For the purposes of taxes on the withdrawals, year 1 of the study will be equated with 1980 in the original table, but the tax rates will equate to the 2014 tax year, with tax calculations (deductions, exemptions and tax brackets) inflated at a 2% rate as previously discussed.
- Both investors will take their first distribution in year 1 when they turn 70 ½ rather than wait until the coming April, to avoid having to take two distributions in the same year.
- Both investors are married and file joint federal tax returns, but no state returns. They also both start taking Social Security at age 70 and have a combined household total Social Security income of $54,000. This increased Social Security income comes from the higher income earner waiting to age 70 to claim a total $36,000 and then adding to that one-half of that total for a spousal income.
- Both couples survive to at least age 99 where this study stops.
- The Social Security income is inflated at a 2% rate.
As you might expect the tax burden on the account grows quite substantially as the retiree gets into his or her eighties and the balance of the tax-deferred account is growing as well. In this study I will look at 5-year data to reduce the size of the charts and data tables. What is interesting to note is the fact that no taxes are due at all from these taxpayers until they reach the age of 76, except for $221 at the age of 73, but after age 76 they start climbing at a steady pace. In the chart below for the RMD strategy, I show the amount of the 5-year totals of RMD income that are used for taxes. Social Security income does not show up in the chart, except that it does contribute to some of the taxes. The reason is that before the addition of the income from the investment account there were no taxes on the $54,000 of income from Social Security; by adding taxable income from the IRA, total after-tax income was increased by the amount shown by the blue bar in the chart below:
The next chart below is for the income-only account and adds together the dividends in the IRA with the dividends generated in the taxable account as the total income available when staying with an income-only strategy. Unfortunately, by age 81 the income-only investor has to start paying taxes on these additional dividends coming from the taxable account because the total taxable income is in the 25% tax bracket and above, so this adds an additional tax burden on the income-only investor. What you can't see from the data provided is that the investor trying to live off of income only at about age 86 and beyond all that income is coming from the taxable account because the dividends in the IRA are completely consumed by taxes. The below chart shows the amount of after-tax income left from the dividends once taxes are paid:
Next I will present a chart that shows the effective tax rates of the two different accounts. The effective tax rate is computed by dividing the total taxes paid by the total income received (Social Security income plus RMD income) for the one account; and dividing the total taxes paid by the Social Security income plus dividend income for the income-only account.
As you can see above, trying to spend only the dividends can leave you in your later years paying almost half of your income to the IRS, from the fact that the RMD withdrawals from your account cause you to pay much more tax than you would otherwise need to pay.
I will now tie this all together with a table that details the income, taxes, and effective tax rates. The first two columns cover account one, which is the income investor that does not want to sell shares, but lives off the income from the dividends only. This investor uses a taxable account for the overflow caused by the RMD withdrawal from the IRA. The third column covers the investor that follows the standard RMD strategy from within an IRA. The last column is a comparison of the income-only investor, if all the money was inside a tax-free Roth account.
One additional note to the above table is about the lack of effective tax rate data for the taxable account. In this case I have created one effective tax rate for the total dividends in both the IRA and taxable account and included that in the column one data. Also note that the income-only investor did not get all of the dividends paid, since a good portion of them were paid in taxes, just slightly over $1 million.
The results of this study surprised even me that in some cases if you think you could live off of the dividends only in an IRA account, the tax bite from a typical dividend growth account could make this virtually impossible with taxes taking close to 50% of your dividend income in later years. Some possible ways to mitigate this problem would be:
- Decide early-on that an income-only strategy with dividend growth stocks is best not attempted from inside an IRA or tax-deferred account where mandatory RMD withdrawals will force you to pay more in taxes.
- Try to spend-down or convert to a Roth your IRA assets early in retirement when tax rates can be controlled to a lower level.
- Use more tax-efficient investments in your taxable account rather than dividend paying stocks, once you get to higher income levels, if you have significant RMD commitments.
Each person must understand their needs and be able to choose the strategy that best fits their needs. Often money may be tight during retirement, and taking on unnecessary taxes may just make the problem worse. In the case of an IRA account it may be preferable to spend-down that account first rather than try to preserve its capital.
It should also not be concluded that just because in this case the total return investor was able to spend significantly more money that this is any kind of guarantee. It is necessary to realize that the amount of money you have to spend during retirement is not only dependent on how much you saved during your accumulation years but also on what you invest in, what type of account the money is in (IRA, Roth, or taxable), and how you withdraw your money during retirement.
You also must realize that past performance is no guarantee of the future, and this applies as much to dividend income as it does to Mr. Market's total returns. These results looked at only one 30-year span of history and by using only one stock the example was set up to mimic a dividend growth portfolio and total return portfolio that had an annualized total return that was a couple percent above the market average. While most dividend investors think this might be entirely possible, your results may turn out to be less than that, in which case your effective tax rates would be lower for the RMD withdrawals.
Disclosure: I have a position in JNJ mentioned in this article. The information provided here is for educational purposes only. It is not intended to replace your own due diligence or professional financial advice.
Disclosure: The author is long JNJ.
The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.