Posts
In a comment to June 19’s post, Anonymous asked what I mean by “pure” annuity. “Pure annuity” is not a term of art, I’m afraid. I made it up. Forgive me for not defining my terms.
What I meant by “pure” annuity is an annuity that pays an agreed annual payment, beginning at an agreed starting date, either for your life, or for the life of you and your spouse. It is “pure” in the sense that it does not contain the following features:
• Payment does not vary with the performance of some asset class, such as with the S&P 500 index; that’s called a “variable” annuity. In my mind, however, an annuity that varies with an inflation index is an acceptably pure annuity.
• Payment is not guaranteed for a minimum period, such as 10 years, should you not live that long. It’s purely for your life. In my mind, however, an annuity that continues for the life of your spouse, should he survive you, is pure enough. Most couples properly view themselves as a single economic unit.
I’m no expert on annuities. I know what I like about them in concept, but I don’t know enough about them to know when they’re a good deal or a rip-off. For a good informative website about annuities, go to AnnuityDigest.com.
Monday, June 29, 2009
Sunday, June 28, 2009
TIPS and Losses
In a comment to June 24’s post, Paul asks a good question: Can “investing in TIPS lead to zero possibility of losses over any and all time periods?”
I think the answer is “no,” but perhaps it depends on the way you look at it. Here are the dangers of TIPS.
Treasury Inflation Protected Securities, or TIPS, are issued by the U.S. government and backed by the full faith and credit of the United States. That means the likelihood of default is close to zero. Not zero, but close to zero. We think of default by the U.S. Treasury as inconceivable, but perhaps it’s more accurate to characterize it as inconceivably bad, rather than inconceivable.
Another obvious danger of TIPS derives from the manner in which it is adjusted for inflation. The Treasury increases the principal value each year to reflect changes in the Consumer Price Index. But the CPI is an imperfect measure of inflation. It’s made up of changes in the prices of a typical basket of goods and services. But a “typical” basket ain’t your basket. The goods and services in the CPI are not necessarily those that you need; nor are they necessarily measured in your part of the country. I don’t foresee the day when the Treasury starts publishing “CPI-Paul.”
Perhaps the most important point to recognize about TIPS and losses is this: Like other bonds, TIPS can gain or lose value in the marketplace. You likely won’t lose principal, but you can lose value. Those are two different things. TIPS are subject to the same laws of supply and demand as other securities. With the government backing of their principal, and their inflation protection, TIPS ought to be less volatile than stocks or other bonds. And I imagine they are, although I have not confirmed that. (Maybe someone reading this post has looked at this. Anyone?) Less volatile, perhaps, but somewhat volatile nonetheless. The market value of TIPS is determined by millions of purchasers and sellers, each acting based on his own financial needs, opinions, and prejudices. So if you have to sell some TIPS to meet your need for cash, you will find that their value has gone up or down, notwithstanding their rather conservative nature. So there is quite a distinct possibility of losses.
You could minimize that possibility by buying TIPS with staggered maturities, so you never have to sell them; just wait around to collect the proceeds as they mature. But your ability to do that is limited by both the impossibility of predicting your future spending needs, and the fact that you simply can’t get TIPS of any maturity in any denomination you wish. They’re not that fine-grained.
So the bottom line is there is no hiding from the possibility of real losses, even with TIPS.
Besides, you would pay a steep price for that low probability of losses in the form of measley returns.
I think the answer is “no,” but perhaps it depends on the way you look at it. Here are the dangers of TIPS.
Treasury Inflation Protected Securities, or TIPS, are issued by the U.S. government and backed by the full faith and credit of the United States. That means the likelihood of default is close to zero. Not zero, but close to zero. We think of default by the U.S. Treasury as inconceivable, but perhaps it’s more accurate to characterize it as inconceivably bad, rather than inconceivable.
Another obvious danger of TIPS derives from the manner in which it is adjusted for inflation. The Treasury increases the principal value each year to reflect changes in the Consumer Price Index. But the CPI is an imperfect measure of inflation. It’s made up of changes in the prices of a typical basket of goods and services. But a “typical” basket ain’t your basket. The goods and services in the CPI are not necessarily those that you need; nor are they necessarily measured in your part of the country. I don’t foresee the day when the Treasury starts publishing “CPI-Paul.”
Perhaps the most important point to recognize about TIPS and losses is this: Like other bonds, TIPS can gain or lose value in the marketplace. You likely won’t lose principal, but you can lose value. Those are two different things. TIPS are subject to the same laws of supply and demand as other securities. With the government backing of their principal, and their inflation protection, TIPS ought to be less volatile than stocks or other bonds. And I imagine they are, although I have not confirmed that. (Maybe someone reading this post has looked at this. Anyone?) Less volatile, perhaps, but somewhat volatile nonetheless. The market value of TIPS is determined by millions of purchasers and sellers, each acting based on his own financial needs, opinions, and prejudices. So if you have to sell some TIPS to meet your need for cash, you will find that their value has gone up or down, notwithstanding their rather conservative nature. So there is quite a distinct possibility of losses.
You could minimize that possibility by buying TIPS with staggered maturities, so you never have to sell them; just wait around to collect the proceeds as they mature. But your ability to do that is limited by both the impossibility of predicting your future spending needs, and the fact that you simply can’t get TIPS of any maturity in any denomination you wish. They’re not that fine-grained.
So the bottom line is there is no hiding from the possibility of real losses, even with TIPS.
Besides, you would pay a steep price for that low probability of losses in the form of measley returns.
Wednesday, June 24, 2009
Stock Market Losses
In a comment to May 7’s post, David asked for some clarification. May 7’s post was about the frequency with which stocks and bonds have exhibited real (i.e., inflation-adjusted) losses over different time periods. It was noted that since 1926, the longest period it has taken stocks to recover their purchasing power after a loss was 19 years. Yes, unfortunately, that was a 19-year stretch; not 19 months. And yes, that means, had you been investing then, you might not have lived long enough to see your investments recover.
Let’s put that observation in perspective (especially since we’ve just been through a major down market in late 2008 and early 2009).
• 19 years is the worst so far. The one we’ve just been through might be even worse yet. History is just history; not a predictor of the future.
• If you’re focusing on worst case, bonds have exhibited even worse results. We have gone through a period where it has taken bonds 52 years to recover their purchasing power. There is no shelter from the storm.
• Take heart! Both stocks and bonds have been winners more often than losers. How much more often? It depends on the length of time you’re looking at, as shown in the graph in May 7’s post. In investing, time is your best friend.
• That particular measure—frequency of a real loss—is just one small aspect of risk and reward. It says nothing about how bad your loss might be. Nor does it say anything about the reward side: How large might you expect your gain to be.
Investment risk and reward is too multi-dimensional to capture in a single measurement. Take a look at March 16’s post for a refresher.
Let’s put that observation in perspective (especially since we’ve just been through a major down market in late 2008 and early 2009).
• 19 years is the worst so far. The one we’ve just been through might be even worse yet. History is just history; not a predictor of the future.
• If you’re focusing on worst case, bonds have exhibited even worse results. We have gone through a period where it has taken bonds 52 years to recover their purchasing power. There is no shelter from the storm.
• Take heart! Both stocks and bonds have been winners more often than losers. How much more often? It depends on the length of time you’re looking at, as shown in the graph in May 7’s post. In investing, time is your best friend.
• That particular measure—frequency of a real loss—is just one small aspect of risk and reward. It says nothing about how bad your loss might be. Nor does it say anything about the reward side: How large might you expect your gain to be.
Investment risk and reward is too multi-dimensional to capture in a single measurement. Take a look at March 16’s post for a refresher.
More on Short-Term Tax Savings vs. Long-term Benefit
Editorial note: This post was supposed to appear on June 23.
Yesterday’s post discussed how there is often a trade-off between short-term tax savings and long-term tax benefit. Paul Anonymous identified one particular manifestation of that issue as he considers whether to contribute to a non-deductible IRA now, or wait until just before he converts it to a Roth IRA in early 2010. The former approach likely has a slightly higher short-term tax cost. The latter approach likely has a slightly higher long-term tax benefit. How do you know which outweighs the other?
Before providing some rules of thumb (which, as we all know, are always wrong), a couple of observations.
Observation #1: You can never know for sure. Accept that, as in all things in life, you must make a decision in the face of uncertainty about what the future will bring. All you can do is make your best guess.
Observation #2: (This one is hard for many people to swallow.) Your guess must necessarily be based on long-term projections. That’s because long-term benefits can only be assessed in the long term. Duh.
Observation #3: Be sure you’re comparing the right output. It’s not which approach pays the least in taxes. Rather, for most of us, it’s which approach leads you to the highest after-tax retirement spending. (For a lucky few—the wealthy—the right output to measure is which approach leads to your children’s highest after-tax wealth.)
So how do you know if you benefit more from a lower short-term tax or from the long-term benefit of a larger Roth IRA? Here’s a couple of rules of thumb:
Rule #1: If you’re 20 or more years away from retirement, then you’ve got enough time to benefit from long-term savings. Pay the higher tax.
Rule #2: If you’re wealthy, regardless of age, then you’ve got enough time to benefit from long-term savings. By “wealthy” I mean either of the following: (1) you have enough assets outside your tax-favored retirement plans to meet your spending needs during your lifetime, and don’t ever need to tap into your tax-favored retirement accounts, except as required by law; or (2) you are spending as much as you care to during your lifetime, and any additional resources will end up in the hands of your beneficiaries after your death. While both categories of "wealthy" benefit from long-term savings, the latter category (the "really wealthy" (?)), derive a greater degree of benefit than the former.
Unfortunately, those two rules of thumb don’t nearly cover the whole waterfront. They still fail to address Paul Anonymous’ situation and that of most readers of this blog. Most of us are not wealthy, and are either at or too near retirement to make use of Rule #1. In that case, the only approach I can recommend is to do a full-blown long-term financial projection. Or, short of that, do what Paul Anonymous is doing and go with your gut instinct.
Yesterday’s post discussed how there is often a trade-off between short-term tax savings and long-term tax benefit. Paul Anonymous identified one particular manifestation of that issue as he considers whether to contribute to a non-deductible IRA now, or wait until just before he converts it to a Roth IRA in early 2010. The former approach likely has a slightly higher short-term tax cost. The latter approach likely has a slightly higher long-term tax benefit. How do you know which outweighs the other?
Before providing some rules of thumb (which, as we all know, are always wrong), a couple of observations.
Observation #1: You can never know for sure. Accept that, as in all things in life, you must make a decision in the face of uncertainty about what the future will bring. All you can do is make your best guess.
Observation #2: (This one is hard for many people to swallow.) Your guess must necessarily be based on long-term projections. That’s because long-term benefits can only be assessed in the long term. Duh.
Observation #3: Be sure you’re comparing the right output. It’s not which approach pays the least in taxes. Rather, for most of us, it’s which approach leads you to the highest after-tax retirement spending. (For a lucky few—the wealthy—the right output to measure is which approach leads to your children’s highest after-tax wealth.)
So how do you know if you benefit more from a lower short-term tax or from the long-term benefit of a larger Roth IRA? Here’s a couple of rules of thumb:
Rule #1: If you’re 20 or more years away from retirement, then you’ve got enough time to benefit from long-term savings. Pay the higher tax.
Rule #2: If you’re wealthy, regardless of age, then you’ve got enough time to benefit from long-term savings. By “wealthy” I mean either of the following: (1) you have enough assets outside your tax-favored retirement plans to meet your spending needs during your lifetime, and don’t ever need to tap into your tax-favored retirement accounts, except as required by law; or (2) you are spending as much as you care to during your lifetime, and any additional resources will end up in the hands of your beneficiaries after your death. While both categories of "wealthy" benefit from long-term savings, the latter category (the "really wealthy" (?)), derive a greater degree of benefit than the former.
Unfortunately, those two rules of thumb don’t nearly cover the whole waterfront. They still fail to address Paul Anonymous’ situation and that of most readers of this blog. Most of us are not wealthy, and are either at or too near retirement to make use of Rule #1. In that case, the only approach I can recommend is to do a full-blown long-term financial projection. Or, short of that, do what Paul Anonymous is doing and go with your gut instinct.
Monday, June 22, 2009
Another Good Roth Conversion Issue
In a comment to June 18’s post, Paul Anonymous has raised another good issue. (Thank you, Paul—whoever you are—you are a veritable fount of good ideas for blog posts.)
Paul’s situation presents a tax-planning issue that comes up all the time, in many sizes and many variations. When is a short-term tax cost outweighed by a long-term benefit? The question is simple. The answer? Not so much.
Here’s Paul’s version of the issue. He plans to make $11,000 nondeductible IRA contribution for 2009 (between him and his spouse, one of whom has reached age 50). He will then convert all his IRAs to Roth IRAs in 2010, when the rules change and the Roth dam bursts. If he contributes the $11,000 now, the tax cost of his Roth conversion will go up on account of any investment earnings between now and the conversion. Paul expects that that Roth conversion tax cost will exceed the tax cost of enjoying those investment earnings outside the shelter of the IRA. Why? Because he intends to invest his $11,000 in stocks, and, at least for the time being, the tax rate on dividends (and long-term capital gain) is lower than the general tax rate that applies to Roth conversions.
(An aside here. Actually, the tax rate on a 2010 Roth conversion might in fact be lower after you factor in the special tax deal on 2010 Roth conversions, as described in February 28’s post. The value of delaying tax by a year and a half brings the effective tax rate down a bit. But for the sake of discussion, let’s assume the tax cost of converting those investment earnings to a Roth account exceeds the tax cost of earning dividends outside the IRA.)
So. Here’s the ubiquitous issue. Is it worth it to pay that extra tax to shift more dollars from your taxable account into your Roth IRA? What’s more powerful—the short-term savings or the long-term benefit?
(Before answering, here’s another aside. The amount at issue here is truly miniscule. In terms of making a meaningful contribution to his future retirement security, Paul is planning to do two big important things with very favorable long-term consequences. (1) He and his wife are contributing the maximum ($11,000 in their case) to their IRAs. (2) And they are planning to convert their traditional IRAs to Roth IRAs in 2010. Way to go, Paul! So why am I even discussing the relatively minor issue of whether he makes the contribution now or just before the Roth conversion? How much could possibly be at stake? $100 investment earnings over the next 7 months maybe? Why am I wasting electronic ink? Just because it’s interesting to me; that’s why. I really need to get a life.)
So, anyway, what’s the answer? Well, it depends on many factors, so there is no one right answer.
But in my experience, after doing untold numbers of projections with varying degrees of precision, I find that the long-term benefit of shifting that hypothetical $100 into the Roth account often exceed the short-term cost of paying a greater rate of tax on an extra $100 in 2010. Or, more precisely, on half in 2011 and half in 2012.
A bit more on this tomorrow.
Paul’s situation presents a tax-planning issue that comes up all the time, in many sizes and many variations. When is a short-term tax cost outweighed by a long-term benefit? The question is simple. The answer? Not so much.
Here’s Paul’s version of the issue. He plans to make $11,000 nondeductible IRA contribution for 2009 (between him and his spouse, one of whom has reached age 50). He will then convert all his IRAs to Roth IRAs in 2010, when the rules change and the Roth dam bursts. If he contributes the $11,000 now, the tax cost of his Roth conversion will go up on account of any investment earnings between now and the conversion. Paul expects that that Roth conversion tax cost will exceed the tax cost of enjoying those investment earnings outside the shelter of the IRA. Why? Because he intends to invest his $11,000 in stocks, and, at least for the time being, the tax rate on dividends (and long-term capital gain) is lower than the general tax rate that applies to Roth conversions.
(An aside here. Actually, the tax rate on a 2010 Roth conversion might in fact be lower after you factor in the special tax deal on 2010 Roth conversions, as described in February 28’s post. The value of delaying tax by a year and a half brings the effective tax rate down a bit. But for the sake of discussion, let’s assume the tax cost of converting those investment earnings to a Roth account exceeds the tax cost of earning dividends outside the IRA.)
So. Here’s the ubiquitous issue. Is it worth it to pay that extra tax to shift more dollars from your taxable account into your Roth IRA? What’s more powerful—the short-term savings or the long-term benefit?
(Before answering, here’s another aside. The amount at issue here is truly miniscule. In terms of making a meaningful contribution to his future retirement security, Paul is planning to do two big important things with very favorable long-term consequences. (1) He and his wife are contributing the maximum ($11,000 in their case) to their IRAs. (2) And they are planning to convert their traditional IRAs to Roth IRAs in 2010. Way to go, Paul! So why am I even discussing the relatively minor issue of whether he makes the contribution now or just before the Roth conversion? How much could possibly be at stake? $100 investment earnings over the next 7 months maybe? Why am I wasting electronic ink? Just because it’s interesting to me; that’s why. I really need to get a life.)
So, anyway, what’s the answer? Well, it depends on many factors, so there is no one right answer.
But in my experience, after doing untold numbers of projections with varying degrees of precision, I find that the long-term benefit of shifting that hypothetical $100 into the Roth account often exceed the short-term cost of paying a greater rate of tax on an extra $100 in 2010. Or, more precisely, on half in 2011 and half in 2012.
A bit more on this tomorrow.
Friday, June 19, 2009
401(k) Investment Options
My friend David asked what I think of adding more participant investment options to his company’s 401(k) plan. He’s on his company’s 401(k) committee, and their investment advisor has recommended adding more options to the current array of 15. Additional fixed income options are on the table.
The easy, short, and largely correct answer is: If that’s what your professional advisor recommends, then do it. That’s what you pay them for.
But maybe David was looking for more substance than that rather facile response.
Initially my reaction is how can more choices ever be bad? If your 401(k) plan is thin on fixed income options, why not add one or two more? It’s always better for participants to have more choices, right?
Not always. Behavioral finance research (which I’m too lazy to look up and cite) has shown that offering too many choices can actually paralyze some people. When faced with too many confusing choices they can react by not choosing at all. In the context of a 401(k) plan, that might mean that their account defaults to some “safe” option determined by the plan, which in the past has often meant a money market fund. In the long run that's a bad choice. (That’s changing, as more and more plans move to a life-cycle fund as the default option.)
For some employees, 401(k) paralysis might be worse. It might mean that they're put off from making an elective deferral at all, which is an awful result.
So the desire to improve the array of options for the group should be tempered by the desire to limit 401(k) paralysis. How many employees can benefit by adding more refined asset classes, compared to how many will now suffer 401(k) paralysis? That’s not an easy question. For one thing, the more fine-grained the asset classes you offer, the fewer employees who can appreciate the subtle distinctions among them. My speculation is that with 15 options you’re getting to the point where only those employees who are receiving professional advice can actually benefit from more options.
Now, having nibbled around the edges of the question, I may as well offer a couple of substantive observations.
One: As long as you’re increasing your plan’s fixed income options, it would be a shame if you fail to offer a TIPS (Treasury Inflation Protected Securities) option.
Two: If you really want to do your employees a favor, offer them a pure annuity option. Not a variable annuity, but a pure annuity that they can add to on a paycheck-by-paycheck basis. This will enable those employees who long for a traditional pension to create one of their own within the confines of your company’s 401(k) plan. (For why I think annuities are valuable for a portion of one's retirement assets, see March 9's post and March 15's post.)
Thank you, David, for giving me the opportunity to shoot my digital mouth off.
The easy, short, and largely correct answer is: If that’s what your professional advisor recommends, then do it. That’s what you pay them for.
But maybe David was looking for more substance than that rather facile response.
Initially my reaction is how can more choices ever be bad? If your 401(k) plan is thin on fixed income options, why not add one or two more? It’s always better for participants to have more choices, right?
Not always. Behavioral finance research (which I’m too lazy to look up and cite) has shown that offering too many choices can actually paralyze some people. When faced with too many confusing choices they can react by not choosing at all. In the context of a 401(k) plan, that might mean that their account defaults to some “safe” option determined by the plan, which in the past has often meant a money market fund. In the long run that's a bad choice. (That’s changing, as more and more plans move to a life-cycle fund as the default option.)
For some employees, 401(k) paralysis might be worse. It might mean that they're put off from making an elective deferral at all, which is an awful result.
So the desire to improve the array of options for the group should be tempered by the desire to limit 401(k) paralysis. How many employees can benefit by adding more refined asset classes, compared to how many will now suffer 401(k) paralysis? That’s not an easy question. For one thing, the more fine-grained the asset classes you offer, the fewer employees who can appreciate the subtle distinctions among them. My speculation is that with 15 options you’re getting to the point where only those employees who are receiving professional advice can actually benefit from more options.
Now, having nibbled around the edges of the question, I may as well offer a couple of substantive observations.
One: As long as you’re increasing your plan’s fixed income options, it would be a shame if you fail to offer a TIPS (Treasury Inflation Protected Securities) option.
Two: If you really want to do your employees a favor, offer them a pure annuity option. Not a variable annuity, but a pure annuity that they can add to on a paycheck-by-paycheck basis. This will enable those employees who long for a traditional pension to create one of their own within the confines of your company’s 401(k) plan. (For why I think annuities are valuable for a portion of one's retirement assets, see March 9's post and March 15's post.)
Thank you, David, for giving me the opportunity to shoot my digital mouth off.
Thursday, June 18, 2009
Roth Conversion Questions
In a comment to May 15’s post, Paul Anonymous raises a number of questions about 2010 Roth IRA conversions. Good man, Paul, you're planning ahead! I’ll try to answer them.
Paul already has both a traditional and a Roth IRA. One question is whether you need to open a separate Roth IRA to hold the assets converted from a traditional IRA, or can you use the existing Roth IRA. The answer is that you don’t need to open a new Roth IRA. One acceptable method for converting a traditional IRA to a Roth is to transfer assets from an existing traditional IRA to an existing Roth IRA. That seems like the administratively easiest thing to do, rather than maintain two separate Roth IRA's needlessly. Although that won’t work if you want to keep the assets at different institutions. Then you’ll of course need separate IRAs.
Paul next asks a question about the wisdom of maintaining a small traditional IRA. Here’s the scenario. Paul apparently intends to convert all (or virtually all) of his traditional IRAs to Roth IRAs. He doesn’t say that explicitly, but that appears to be his plan. And apparently he does not expect to meet the qualification requirements for annual Roth IRA contributions—too much Adjusted Gross Income perhaps. So, looking to the future—2011 and beyond—Paul is planning to make annual traditional IRA contributions and then immediately turn around and convert them to Roth IRAs. Good planning! But that raises the question of whether he should leave some money in a traditional IRA—perhaps a nominal amount—so he doesn’t have to re-open a new traditional IRA every year. I think that’s a terrific idea, as it saves the administrative headache of opening a new account every year--a process that has gotten increasingly burdensome. Keep just enough in the traditional IRA to minimize administrative fees, and then convert the rest to a Roth IRA.
Paul plans to delay his 2009 nondeductible traditional IRA contribution until early 2010. At that point, he will make both his 2009 and 2010 nondeductible traditional IRA contributions, and then immediately turn around and convert them to Roth IRAs. Paul’s thinking is that by delaying the 2009 contribution, he will avoid investment earnings on that amount and reduce the tax cost of converting his whole traditional IRAs to Roth IRAs.
Not so fast, Paulie. I’ve got to disagree with your analysis there. If you keep your $5,000 in your taxable account for the rest of 2009, aren’t you going to earn some returns inside your taxable account—say $50? And then won’t that therefore generate an income tax. In fact, if you invest in interest-bearing investments, the tax you save on the Roth conversion will exactly equal the tax you’ll owe on your taxable investment earnings, so there will be no tax savings at all, plus you will have lost the opportunity to house the $50 inside that most valuable Roth investment environment. Even if you invest in equities, the earnings on which enjoy lower tax rates, the long run benefits of having the $50 inside the Roth environment easily outweigh the short-term detriment of paying a greater Roth conversion tax than a corresponding income tax on the $50 investment earnings. So if I were Paul, I would make my 2009 traditional IRA contribution now, and thereby divert that hypothetical $50 from my taxable bucket to my tax-exempt bucket.
An aside here. It appears that Paul is planning to convert virtually all his traditional IRAs to Roth IRAs in 2010. Would it still be his best strategy to make a 2009 nondeductible IRA contribution if he were planning to convert only a portion of his IRA's to Roths—let’s say 25% of his traditional IRAs? That’s a tougher question. Because then he would only get to recover 25% of his nondeductible contributions tax-free in 2010, and he’d have to wait until later years to recover the remaining 75%. Let’s say he makes a $5,000 nondeductible traditional IRA contribution in 2009, and then converts 25% of his traditional IRAs to a Roth in 2010. The amount he pays tax on is reduced by 25% of his cumulative non-deductible contributions, including the one from 2009. So he contributes $5,000—no deduction—and ends up paying a Roth conversion tax on $3,750. He’s just increased his taxable income by $3,750; he’ll eventually recover the other $3,750 tax-free, but it will take years for that to happen. Is that good planning? I don’t know. I suspect it is, but I think the issue needs some closer study. Look for more on this question in a future post.
Paul already has both a traditional and a Roth IRA. One question is whether you need to open a separate Roth IRA to hold the assets converted from a traditional IRA, or can you use the existing Roth IRA. The answer is that you don’t need to open a new Roth IRA. One acceptable method for converting a traditional IRA to a Roth is to transfer assets from an existing traditional IRA to an existing Roth IRA. That seems like the administratively easiest thing to do, rather than maintain two separate Roth IRA's needlessly. Although that won’t work if you want to keep the assets at different institutions. Then you’ll of course need separate IRAs.
Paul next asks a question about the wisdom of maintaining a small traditional IRA. Here’s the scenario. Paul apparently intends to convert all (or virtually all) of his traditional IRAs to Roth IRAs. He doesn’t say that explicitly, but that appears to be his plan. And apparently he does not expect to meet the qualification requirements for annual Roth IRA contributions—too much Adjusted Gross Income perhaps. So, looking to the future—2011 and beyond—Paul is planning to make annual traditional IRA contributions and then immediately turn around and convert them to Roth IRAs. Good planning! But that raises the question of whether he should leave some money in a traditional IRA—perhaps a nominal amount—so he doesn’t have to re-open a new traditional IRA every year. I think that’s a terrific idea, as it saves the administrative headache of opening a new account every year--a process that has gotten increasingly burdensome. Keep just enough in the traditional IRA to minimize administrative fees, and then convert the rest to a Roth IRA.
Paul plans to delay his 2009 nondeductible traditional IRA contribution until early 2010. At that point, he will make both his 2009 and 2010 nondeductible traditional IRA contributions, and then immediately turn around and convert them to Roth IRAs. Paul’s thinking is that by delaying the 2009 contribution, he will avoid investment earnings on that amount and reduce the tax cost of converting his whole traditional IRAs to Roth IRAs.
Not so fast, Paulie. I’ve got to disagree with your analysis there. If you keep your $5,000 in your taxable account for the rest of 2009, aren’t you going to earn some returns inside your taxable account—say $50? And then won’t that therefore generate an income tax. In fact, if you invest in interest-bearing investments, the tax you save on the Roth conversion will exactly equal the tax you’ll owe on your taxable investment earnings, so there will be no tax savings at all, plus you will have lost the opportunity to house the $50 inside that most valuable Roth investment environment. Even if you invest in equities, the earnings on which enjoy lower tax rates, the long run benefits of having the $50 inside the Roth environment easily outweigh the short-term detriment of paying a greater Roth conversion tax than a corresponding income tax on the $50 investment earnings. So if I were Paul, I would make my 2009 traditional IRA contribution now, and thereby divert that hypothetical $50 from my taxable bucket to my tax-exempt bucket.
An aside here. It appears that Paul is planning to convert virtually all his traditional IRAs to Roth IRAs in 2010. Would it still be his best strategy to make a 2009 nondeductible IRA contribution if he were planning to convert only a portion of his IRA's to Roths—let’s say 25% of his traditional IRAs? That’s a tougher question. Because then he would only get to recover 25% of his nondeductible contributions tax-free in 2010, and he’d have to wait until later years to recover the remaining 75%. Let’s say he makes a $5,000 nondeductible traditional IRA contribution in 2009, and then converts 25% of his traditional IRAs to a Roth in 2010. The amount he pays tax on is reduced by 25% of his cumulative non-deductible contributions, including the one from 2009. So he contributes $5,000—no deduction—and ends up paying a Roth conversion tax on $3,750. He’s just increased his taxable income by $3,750; he’ll eventually recover the other $3,750 tax-free, but it will take years for that to happen. Is that good planning? I don’t know. I suspect it is, but I think the issue needs some closer study. Look for more on this question in a future post.
Wednesday, June 17, 2009
Keeping Track of Nondeductible IRA Contributions
In a comment to Monday’s post, David raises a question about keeping track of non-deductible contributions to a traditional IRA. He asks if his IRA manager should be expected to keep track of this for him.
The short answer is “no.” Your IRA trustee or custodian—or for that matter, any other financial institution involved with your IRA, such as an investment advisor or mutual fund company—is not obligated to keep track of your nondeductible contributions. Not only are they not obligated to do this, they can’t do it, since the facts that determine whether a given IRA contribution is deductible or not is beyond their knowledge. They don’t know about your status as an active participant in your employer’s retirement plan; nor do they know your AGI. So if you think you’ve lost track of your nondeductible contributions, you can’t look to them for help.
(By the way, the rules for determining whether a traditional IRA contribution is deductible or not—which are way more complicated than they need to be—are summarized in February 14’s post.)
But wait! All is not lost! The only one who can keep track of your cumulative nondeductible traditional IRA contributions is you. And chances are pretty good you’ve done so, even if you don’t remember. Here’s why. Whenever you make a nondeductible traditional IRA contribution, you’re required to file a Form 8606 with your tax return for that year. And that form picks up your cumulative nondeductible contributions from prior years and makes you keep a running total. So all you have to do is find your tax return from the most recent year in which you made a nondeductible contribution, and, voila!, there it is. You don’t have to comb through all prior years’ returns—just the most recent one.
And then when you start taking distributions later in life, the same Form 8606 helps you (actually, makes you) keep track of how much of your distribution is tax-free, and how much you have left to recover tax-free in later years.
If you can’t find your tax return for the year you last made a nondeductible contribution, all is not lost. Maybe your accountant kept a copy. Or if you used tax preparation software, maybe the software kept track of it on your computer. I know from personal experience that Turbo-Tax keeps a running record of it, and carries it forward from year to year.
One other point David raised that I want to touch on. In his comment he asks if he should have expected his IRA manager to have “segregated” his nondeductible contributions. Segregating your nondeductible contributions is neither necessary nor at all helpful. That’s because, as described in February 15’s post, when it comes time to take distributions and enjoy some tax-free distributions, IRS rules force you to aggregate all your traditional IRAs in figuring the amount that’s tax-free. Segregating them into one IRA does you no good.
’Til manana.
The short answer is “no.” Your IRA trustee or custodian—or for that matter, any other financial institution involved with your IRA, such as an investment advisor or mutual fund company—is not obligated to keep track of your nondeductible contributions. Not only are they not obligated to do this, they can’t do it, since the facts that determine whether a given IRA contribution is deductible or not is beyond their knowledge. They don’t know about your status as an active participant in your employer’s retirement plan; nor do they know your AGI. So if you think you’ve lost track of your nondeductible contributions, you can’t look to them for help.
(By the way, the rules for determining whether a traditional IRA contribution is deductible or not—which are way more complicated than they need to be—are summarized in February 14’s post.)
But wait! All is not lost! The only one who can keep track of your cumulative nondeductible traditional IRA contributions is you. And chances are pretty good you’ve done so, even if you don’t remember. Here’s why. Whenever you make a nondeductible traditional IRA contribution, you’re required to file a Form 8606 with your tax return for that year. And that form picks up your cumulative nondeductible contributions from prior years and makes you keep a running total. So all you have to do is find your tax return from the most recent year in which you made a nondeductible contribution, and, voila!, there it is. You don’t have to comb through all prior years’ returns—just the most recent one.
And then when you start taking distributions later in life, the same Form 8606 helps you (actually, makes you) keep track of how much of your distribution is tax-free, and how much you have left to recover tax-free in later years.
If you can’t find your tax return for the year you last made a nondeductible contribution, all is not lost. Maybe your accountant kept a copy. Or if you used tax preparation software, maybe the software kept track of it on your computer. I know from personal experience that Turbo-Tax keeps a running record of it, and carries it forward from year to year.
One other point David raised that I want to touch on. In his comment he asks if he should have expected his IRA manager to have “segregated” his nondeductible contributions. Segregating your nondeductible contributions is neither necessary nor at all helpful. That’s because, as described in February 15’s post, when it comes time to take distributions and enjoy some tax-free distributions, IRS rules force you to aggregate all your traditional IRAs in figuring the amount that’s tax-free. Segregating them into one IRA does you no good.
’Til manana.
Tuesday, June 16, 2009
The Value of Budgetary Flexibility
Here’s a question for you:
If you said, “E. All of the above,” shame on you. You’re not being responsive to the question. I said greatest impact. Of course they’re all important, but you have to pick one.
My educated guess is that the answer is D. Flexibility in your budget. Someday I hope to do a study confirming the relative importance of those factors, but for now it remains an educated guess. Consider the following tale.
Rita Rigid and Zelda Zen are twin sisters. Both are about to retire with $2,000,000 of assets (in a traditional IRA). Both use the same financial advisor. Both want a standard of living that’s as luxurious as possible, consistent with prudence—neither wants to run out of money. The sisters are alike in every way, except for one key difference. While Rita Rigid wants to live on a budget that doesn’t fluctuate from year to year, except for increasing with inflation, Zelda Zen believes she can stand some ups and downs in her annual spending.
Here is the advice they get after consulting with their financial advisor. Rita is advised to plan on spending $70,000 per year, increased annually by inflation, plus necessary income taxes. That’s 3.5% of her starting assets. Zelda is advised to spend $110,000 plus necessary income taxes. Every year, Zelda's budget is to be 5.5% of her remaining assets, but no less than $60,000 and no more than $130,000 (with those brackets increased for inflation each year). Zelda is told to expect her future budget to be both unknown and erratic, to fluctuate significantly between the $60,000 and $130,000 brackets. But the central tendency of her expected annual budget will be about $96,000 (plus taxes).
Depending on which number you look at (Zelda’s $110,000 starting budget or her $96,000 expected budget), Zelda’s standard of living is either 57% or 37% greater than Rita’s. That entire difference arises from her willingness to be flexible in her annual spending. Of course Zelda might experience an unlucky investment environment in the future, and her spending will be less than Rita’s; but she has expressed a willingness to accept that. And it is precisely that willingness that enables her to prudently spend more than Rita.
These figures come from an article I wrote that will appear in a professional journal later this year (CCH Journal of Retirement Planning). They were not just pulled out of a hat, but were based on a rigorous study of how Rita’s and Zelda’s investments might be expected to behave in uncertain markets. Rita prudently deals with that uncertainty by reducing her spending to a conservative amount so that she has a low likelihood of running out of money. Her desire for budgetary certainty demands that conservatism. Zelda prudently deals with uncertainty by anticipating that her spending will fluctuate—within the limits she has set—as the unknowable future unfolds.
Either approach is legitimate. But it’s eye-opening to see that Zelda’s budgetary flexibility is worth a whopping 37%-57% of her standard of living. It’s positively Zen-like. The less you must spend, the more you can spend.
Which of the following has the greatest impact on your standard of living in retirement?
A. Employing a wise asset allocation.
B. Having a wise investment advisor.
C. Having a wise tax advisor.
D. Having flexibility in your budget.
If you said, “E. All of the above,” shame on you. You’re not being responsive to the question. I said greatest impact. Of course they’re all important, but you have to pick one.
My educated guess is that the answer is D. Flexibility in your budget. Someday I hope to do a study confirming the relative importance of those factors, but for now it remains an educated guess. Consider the following tale.
Rita Rigid and Zelda Zen are twin sisters. Both are about to retire with $2,000,000 of assets (in a traditional IRA). Both use the same financial advisor. Both want a standard of living that’s as luxurious as possible, consistent with prudence—neither wants to run out of money. The sisters are alike in every way, except for one key difference. While Rita Rigid wants to live on a budget that doesn’t fluctuate from year to year, except for increasing with inflation, Zelda Zen believes she can stand some ups and downs in her annual spending.
Here is the advice they get after consulting with their financial advisor. Rita is advised to plan on spending $70,000 per year, increased annually by inflation, plus necessary income taxes. That’s 3.5% of her starting assets. Zelda is advised to spend $110,000 plus necessary income taxes. Every year, Zelda's budget is to be 5.5% of her remaining assets, but no less than $60,000 and no more than $130,000 (with those brackets increased for inflation each year). Zelda is told to expect her future budget to be both unknown and erratic, to fluctuate significantly between the $60,000 and $130,000 brackets. But the central tendency of her expected annual budget will be about $96,000 (plus taxes).
Depending on which number you look at (Zelda’s $110,000 starting budget or her $96,000 expected budget), Zelda’s standard of living is either 57% or 37% greater than Rita’s. That entire difference arises from her willingness to be flexible in her annual spending. Of course Zelda might experience an unlucky investment environment in the future, and her spending will be less than Rita’s; but she has expressed a willingness to accept that. And it is precisely that willingness that enables her to prudently spend more than Rita.
These figures come from an article I wrote that will appear in a professional journal later this year (CCH Journal of Retirement Planning). They were not just pulled out of a hat, but were based on a rigorous study of how Rita’s and Zelda’s investments might be expected to behave in uncertain markets. Rita prudently deals with that uncertainty by reducing her spending to a conservative amount so that she has a low likelihood of running out of money. Her desire for budgetary certainty demands that conservatism. Zelda prudently deals with uncertainty by anticipating that her spending will fluctuate—within the limits she has set—as the unknowable future unfolds.
Either approach is legitimate. But it’s eye-opening to see that Zelda’s budgetary flexibility is worth a whopping 37%-57% of her standard of living. It’s positively Zen-like. The less you must spend, the more you can spend.
Monday, June 15, 2009
Losses in Retirement Accounts
I’m back from vacation! Actually I was back a couple of weeks ago, but I was a bit lazy about getting back to The Two Legged Stool. But I’m here now.
I thought I would restart by responding to a couple of comments. First Paul. In his comment to the May 15 post, Anonymous (aka Paul) raises a couple of issues. One of them deals with losses inside a nondeductible IRA. Paul wonders if you can deduct up to $3,000 of losses that you have incurred inside your nondeductible Individual Retirement Account.
You can’t directly deduct losses that occur inside your IRA. Instead, you indirectly get the tax benefit of your losses by having that much less to distribute during your retirement years and therefore reporting that much less taxable income. Remember, everything that goes on inside your IRA is generally irrelevant for income tax purposes. You don’t (yet) pay taxes on interest or dividends; you don’t pay taxes on any capital gains; and you don’t get to deduct any capital losses.
There is one possibility for deducting a tax loss in a nondeductible IRA, but it’s very unusual and I hope it never happens to you. Here it is. When you are at the point of taking out the last dollars from all your traditional IRAs, deductible and nondeductible, if you still have any nondeductible contributions left that you haven’t yet deducted (also sometimes called “basis”), and your losses have been so great over your lifetime that your basis actually exceeds the amount remaining in your IRA, then you can take a tax loss for the difference. I hope this doesn’t happen to you! Picture how unusual it is. It could only happen at the waning years of your life when you’re ready to completely deplete your IRA. (It could also happen sooner if you convert all your traditional IRAs to Roth IRAs.) And your losses would have had to have been pretty bad to exceed all your income and gains during your lifetime. And you have to look at all your traditional IRAs in the aggregate when figuring whether you have a net loss or not. So overall it’s pretty unusual to realize an actual net deductible loss on your IRA.
It's good to be back.
I thought I would restart by responding to a couple of comments. First Paul. In his comment to the May 15 post, Anonymous (aka Paul) raises a couple of issues. One of them deals with losses inside a nondeductible IRA. Paul wonders if you can deduct up to $3,000 of losses that you have incurred inside your nondeductible Individual Retirement Account.
You can’t directly deduct losses that occur inside your IRA. Instead, you indirectly get the tax benefit of your losses by having that much less to distribute during your retirement years and therefore reporting that much less taxable income. Remember, everything that goes on inside your IRA is generally irrelevant for income tax purposes. You don’t (yet) pay taxes on interest or dividends; you don’t pay taxes on any capital gains; and you don’t get to deduct any capital losses.
There is one possibility for deducting a tax loss in a nondeductible IRA, but it’s very unusual and I hope it never happens to you. Here it is. When you are at the point of taking out the last dollars from all your traditional IRAs, deductible and nondeductible, if you still have any nondeductible contributions left that you haven’t yet deducted (also sometimes called “basis”), and your losses have been so great over your lifetime that your basis actually exceeds the amount remaining in your IRA, then you can take a tax loss for the difference. I hope this doesn’t happen to you! Picture how unusual it is. It could only happen at the waning years of your life when you’re ready to completely deplete your IRA. (It could also happen sooner if you convert all your traditional IRAs to Roth IRAs.) And your losses would have had to have been pretty bad to exceed all your income and gains during your lifetime. And you have to look at all your traditional IRAs in the aggregate when figuring whether you have a net loss or not. So overall it’s pretty unusual to realize an actual net deductible loss on your IRA.
It's good to be back.
Subscribe to:
Posts (Atom)