Tuesday, December 31, 2013

Why Bonds?

Wade Pfau recently posted a nice piece on how to build a TIPs ladder for retirement. A discussion ensued on the topic of whether one should build a ladder of individual TIPs bonds or instead buy a fund of TIPs bonds.

I'll chime in on the topic, but first let's talk about why a retiree should own bonds at all.

There are three basic alternatives for investing your savings after retirement. You can buy a life annuity from an insurance company that will pay you periodic amounts (let's assume monthly) for as long as you live. You will continue to receive these payments if you live to age 150, but you will stop receiving them when you die, even if that's next year. 

The best thing about a life annuity is that you will never run out of money. The worst thing may be that you could end up with nothing left for your heirs.

(There are options you can buy to protect yourself against losing your investment if you don't live at least 10 years, to continue payments to a surviving spouse, and you can even purchase inflation protection, but let's not get into the weeds at this point.)

The second major alternative is to buy bonds. As William Sharpe and Jason Scott showed in "A 4% Rule -- At What Price?", you can invest in a ladder of TIPs bonds and, if interest rates follow long-term average returns of 2%, you can spend 4.46% of your initial portfolio value every year and your capital should last exactly 30 years. In other words, you can withdraw a constant $44.60 adjusted for inflation every year for 30 years for every $1,000 invested.

There are at least two major differences between these alternatives. First, the TIPs ladder will last exactly thirty years, at which time your account balance will be zero.  The annuity pays for the remainder of your life, which could be significantly more or less than 30 years.

Second, you give your principal to the insurance company up front for an annuity, but you always own the bonds in your bond ladder. If you live less than 30 years, you can leave the surplus bonds to heirs. Depending on the options you choose, there may be nothing left of an annuity to bequeath.

The third major alternative for your post-retirement investment dollars is a stock portfolio. You could invest in stocks and spend from the portfolio each year. Maybe you could pay your annual expenses and end up with a large portfolio to leave your heirs. Or maybe you will completely run out of money long before you die. There's a lot of upside potential with this approach, and a roughly equal downside.

A commenter on the Pfau thread suggested that he prefers investing in dividend-generating stocks, with a goal of spending dividends of around 4% and, unlike the annuity or TIPs ladder approach, being able to preserve his capital. Preserving capital is, in fact, one possible outcome. Going broke in old age is another. Stocks don't always go up.

Another commenter on the Pfau thread asked why you would invest in risky stocks and spend 4% a year when you could invest in a TIPs ladder and spend 4.5%. Part of the answer is that at the end of 30 years, the TIPs ladder is completely spent, principal and all. The value of the stock portfolio, on the other hand, could be enormous after 30 years, or it might not last 20 years.

The other part of the answer is that 4.5% is a pretty predictable spend rate for the TIPs ladder, while 4% for the stock portfolio is merely a guess.

Of course, you can bet some of your retirement on a combination of two or three of these alternatives, and that is probably the more common strategy.

So, back to why a retiree should own bonds. If you decide to go the stock portfolio route, you should probably also own some bonds. As Modern Portfolio Theory predicts, bonds can decrease the risk of a portfolio a lot while lowering its return just a little. Deciding how much of your portfolio should be held in bonds at what age is still hotly debated.
(from Young Research & Publishing)
As an example, Index Fund Advisors calculates that the long term average return for a portfolio of 50% stocks and 50% bonds is 8.15% with a standard deviation (risk) of 11.42%. Lowering the stock allocation from 50% to 40% reduces the expected return to 7.39% (9.33% lower) but reduces the standard deviation to 9.28% (an 18.7% reduction of risk).

Another reason to own bonds is that they can provide a safe, predictable amount of future income. Let's say you predict that you will need $30,000 in 2019, five years from now, and you can find TIPs bonds in the market that mature in 2019 that currently offer a real yield-to-maturity of 2%. Let's simplify matters by assuming that the TIPs bond you find is a zero coupon bond (there aren't any, for some reason). If you invest $27,172 in such bonds, you can be pretty sure that the bond will be worth $30,000 in 2013 dollars when it matures in 2019.

Do that for the next 30 years (or any number of consecutive years) and you have a bond ladder. You also have a safe, predictable, inflation-protected income stream.

Why buy bonds, then? Because they improve your risk/reward profile if you decide to go the stocks route and they can provide safe, predictable, inflation-protected income if you decide to go with a TIPs ladder. Remember these two benefits, because which you desire will be a determinant of whether you should buy a fund or individual TIPs. More on that later.

Go the annuity route and you probably don't need bonds. In fact, a fixed annuity is a lot like a bond, issued by an insurance company, with a lifetime coupon and no remaining value when you die.

Unless you annuitize all your retirement savings, you're probably going to want to own some bonds to reduce your stock portfolio volatility, or to ensure income for living expenses for some future years.

Probably both.

Wade's original question, though, was whether to build a TIPs ladder or to buy a fund, but we're not there, yet. Now that we've discussed why to buy bonds at all, the next question is "Why TIPs bonds"?

Sunday, December 29, 2013

Saving Excess Tax Deductions at the Last Minute

I’m not a tax specialist. I have my taxes done by an expert IRS-enrolled agent. Keeping track of tax laws is a full time job and one I'm delighted to outsource.

I do have a tax idea to pass along that sometimes works for retirees. Please discuss this with a tax expert if you’re interested. 

Retirees can find themselves in the position of not only owing no federal taxes (because they’re no longer working), but actually having more deductions than they can use. Health care and health insurance costs, for example, can generate far more tax deductions than you might have considered when you were budgeting for retirement. 

Health care costs have been the second largest financial shock of my retirement (right after learning that kids don’t actually grow up, go to college and then support themselves, anymore).

This may especially be the case if you’re spending first from your taxable accounts, and thus not generating taxable income from IRA’s and 401(k)’s. You might just find that you have no taxable federal income and, in fact, you have more itemized deductions than you can use.

Here's an example. Joe retired at the beginning of 2013 and paid his living expenses from his taxable retirement savings. $10,000 interest and dividends from his taxable accounts makes up his only taxable income. He has no taxable income from Social Security (he hasn't claimed, yet) and no taxable withdrawals from IRA's or 401(k) accounts.

Joe has $30,000 in mortgage interest and deductible property taxes and paid another $30,000 for deductible medical costs and private health insurance, and of course, exemptions for himself and his spouse.

According to the TurboTax online tax estimator, Joe should expect to report $10,000 of taxable income, $59,250 deductions and $15,600 exemptions. He should expect to pay no federal taxes.

No federal taxes sounds good, but Joe could have had another $65,000 of taxable income and still paid no federal taxes because he has unused tax deductions.

Now, you can carry forward capital gains losses and net operating losses to future years when you do owe taxes, but you can’t carry forward unused itemized deductions.

So, how can a retiree make use of all those deductions? By generating more income. And, how do you do that the last week of the year when you didn’t do it for the previous 51?

Convert a traditional IRA to a Roth. The conversion amount will be reported on IRS Form 8606 and claimed as an IRA withdrawal in the Income section near the top of IRS Form 1040. The amount will be taxed as regular income, so the trick is to convert the amount of your IRA account to a Roth account that will be offset by your excess deductions and leave your taxes at zero, or thereabouts.

The TurboTax estimator tells us that if Joe converts $75,000 from his traditional IRA to a Roth IRA, increasing his taxable income from $10,000 to $75,000, he will still pay no taxes. (The estimator says $4, but let's not quibble.)

You don't pay income tax on Roth IRA withdrawals, but traditional IRA withdrawals are taxed as ordinary income. Leave the money in an IRA and you will eventually be taxed when you take it out. The funds you convert to a Roth will no longer be subject to income tax when you spend them and you will have achieved this by paying the current taxes dues on the IRA withdrawal, which, if you do it right, will be zero.

The amount you convert doesn’t need to be exact because, should you convert too much, you can undo all or part of the conversion in 2014 with a Roth recharacterization. I suggest you estimate your taxes, guess the amount of excess deductions you will have, and convert that amount or a little more. (You can put money back into your IRA in 2014, but you can’t take out more.)

Let's say Joe converts $75,000 by the end of 2013 but his tax preparer points out another $10,000 in taxable income that Joe didn't realize would be treated as taxable income. Joe can simply put $10,000 back in his IRA through a recharacterization.

As I said, you can undo all of this in 2014, but the conversion has to be done by the end of 2013 to offset taxes this year.

There are deadlines for the recharacterization, the earliest of which is the day your 2013 taxes are due but this can be extended to October 15.

What happens if Joe only converts $30,000 and finds out after his taxes are prepared next April that he could have converted $75,000? Too bad. You can put converted dollars back after the end of the tax year, but you can't take out more.

Of course, there are many reason to do a Roth conversion, or not to, and retirees aren't the only potential beneficiaries. Maybe you were unemployed for part of 2013, but still had high medical bills and deductible mortgage interest. You still might find yourself with more deductions than you can use.

I can’t say this enough: I am not a tax expert. Roth conversions, like many tax maneuvers, have subtleties that need to be considered.  I am not suggesting that you do this, I am suggesting that you might want to discuss this with a tax expert.

You'll have to act quickly to do it this year. You have until close of business Tuesday to convert. It 's pretty easy at websites like Fidelity's. It takes just a few minutes.

Of course, there's always next December.

Happy New Year to all my fellow (and future) retiree readers.

Wednesday, December 4, 2013

Does Retirement Get Cheaper as We Age?

A reader noted last week that living expenses tend to decline as we age and suggested that tendency offers some comfort as we look at the cost of a long retirement. That is true in one respect, but there are a couple of important considerations to keep in mind.

First, let’s look at the true part. Expenses typically decline with age, even when we include health care costs.

According to a 2012 study, Expenditure Patterns of Older Americans, 2001‒2009 by Sudipto Banerjee, Employee Benefit Research Institute, "Household expenses steadily decline with age. With the age 65 expenditure as a benchmark, household expenditure falls by 19 percent by age 75, 34 percent by age 85, and 52 percent by age 95.

Here’s a chart of household expenditures by age from the study1. It's almost linear.
What doesn’t get cheaper as we age? Health care. The cost of every other category in the study (home, food, clothing, etc.) decreased with age.

According to Banerjee, “Health-related expenses are the second-largest component in the budget of older Americans. It is the only component which steadily increases with age. Health care expenses capture around 10 percent of the budget for those between 50–64, but increase to about 20 percent for those age 85 and over.

Health care expenses probably won't steadily decline with age. They're more likely to come in bunches. Studies have shown that many retirees will spend most of their total lifetime medical expenses near the end of their life.

So, yes, living expenses tend to decline as we age but growing healthcare expenses offset those savings to some degree. 

The real wild card among retirement expenses is long term care. 

It is estimated that about 60% of Americans will need some form of long term care in their lifetime and this care is not covered by Medicare. That’s the scary number LTC insurance providers use to sell insurance policies but, in reality, the costs of long term care range from several thousand dollars that might be easily covered out of pocket all the way up to "catastrophic".

As I showed in a previous blog on Long Term Care Insurance, more than 40% of retiree’s can expect no long term care costs at all and another 19% can expect costs less than $10,000. An additional 8% can expect costs between $10,000 and $25,000. So, arguably about 70% of retirees will have LTC costs that might well be managed without insurance.

Still, that leaves 30% of retirees that can expect to have very large, perhaps catastrophic long term care costs.

The following chart shows the percentages for those who will need long term care at various cost levels.
You may not have long term care needs at all, or you may only need care for a short time and experience manageable costs, but the possibility remains that you or your spouse might have catastrophic long term care needs that will destroy your financial plans for late retirement. 

LTC insurance is available, but costly. Wealthy people can self-insure. Those with scarce resources won’t be able to afford it. That leaves a group in the middle with a decision to make about purchasing insurance.

There is another major unknown factor in the cost of our retirement, of course: how long we will live. A shorter retirement will cost less, perhaps far less, than one that takes us into our late nineties. While you might derive some consolation from the fact that expenses tend to decline as we age, the factor that really increases or decreases the cost of retirement is longevity. 

Given the possibilities of a very long life and catastrophic LTC costs, expenses in retirement can be wildly unpredictable. Part of that risk depends on how you choose to finance it. Those who choose to forgo (or cannot afford) LTC insurance and choose to fund retirement with a stock and bond portfolio will have little control over those risks. The range of potential outcomes is quite large.

Households that insure against LTC and longevity risk by purchasing LTC insurance and life annuities will have less money to spend elsewhere, but will avoid the worst-case outcomes and have more predictable expenses.

So, will your total expenses decline as you age? If you stay healthy they probably will.

If you have high health care costs and uninsured long term care costs, expenses can explode.

1A lot of retirement income calculators treat income and expenses as static throughout retirement. "You need to replace 70% of your pre-retirement income." As this chart shows, expenses are not static, and income isn't typically static, either. A good retirement plan takes these changes over time into consideration. I prefer a consumption-smoothing tool like E$Planner.