Substantially Equal Periodic Payments and

Other Strange Animals

Found in the

Internal Revenue Code

by

William J. Stecker, CPA

Second Edition © 2001

Extremely Important Notice The second edition of this text was published November 2001. From that date through October 4, 2002, everything in the text was accurate. However, on October 4, 2002, the Internal Revenue Ser-vice issued Revenue Ruling 2002-62. Although only a seven-page document, this ruling is extremely significant and literally creates a whole new set of rules regarding both in-progress as well as new sub-stantially equal periodic payment plans. In general, this new ruling is two rulings in one:

(1) It provides for new interpretations of IRC §72(t)(4), specifically what constitutes a “modification” and what does not; thus, providing a relief mechanism to taxpayers with existing SEPP plans that are prematurely running out of IRA assets. Further, a one-time method conversion is allowed to the “required minimum distribution” method.

(2) Effective, January 1, 2003, new SEPP plans will be materially limited from the current

set of rules such that the maximum amount of distributions allowed, holding the IRA balance constant, will be reduced in the range of 25% to 30%.

As a result of the new ruling, this text is undergoing a major revision which is going to take some time to accomplish—both in resolving new open issues related to the new ruling as well as a complete re-write of the text itself. Anyone purchasing this report should do several things: one, send an e-mail to themarblegroup@earthlink.net so that the author can send you, at no charge, a copy of the 3rd edition when it becomes available (anticipated in January 2003); two, should you anticipate launching a new SEPP program in the near future, please seek professional assistance from the tax accountant or tax attorney of your choice, at least, to get briefed on the contents of Revenue Ruling 2002-62 and how it may change your SEPP plans. As an interim source of information, feel free to visit www.72t.net and review both the Articles section of the web site as well as the Message Board. There are several good articles here as well as ongoing discussion of tactical SEPP issues. William J. Stecker, CPA

iii

Table of Contents

Introduction ..............................................................................................................vi Second Edition ....................................................................................................... viii Disclaimer............................................................................................................... viii Biographical Information...................................................................................... viii Chapter 1: Current Law............................................................................................1 Internal Revenue Code................................................................................................................1 Exception #1 Age 59½................................................................................................................1 Exception #2 Death.....................................................................................................................1 Exception #3 Disability...............................................................................................................2 Exception #5 Separation of Service at Age 55.............................................................................4 Exceptions #6 Through #12 ........................................................................................................5 Exception #4 Substantially Equal Period Payments .....................................................................7 Internal Revenue Service Publications ........................................................................................9 Tax Court Cases..........................................................................................................................9 Private Letter Rulings .................................................................................................................9 Chapter 2: Common Concepts ................................................................................11

Unearned Ordinary Income ....................................................................................................... 11 Age Windows ...........................................................................................................................11 The Account Concept................................................................................................................12 Individual Ownership................................................................................................................13 Relationship to Earned Income..................................................................................................14 Reversibility and Errors ............................................................................................................14 Basis in Account .......................................................................................................................15 Penalty Computations ...............................................................................................................15

Chapter 3: Other Strange Animals—Roth IRAs & NUA......................................17 Roth IRA Taxation and Penalties ..............................................................................................17 Roth Definition .........................................................................................................................17 Qualified and Unqualified Withdrawals ....................................................................................17 Application of 5-Taxable-Year Rule .........................................................................................18 Contribution and Earnings Withdrawals ....................................................................................19 Conversion Dollar Withdrawals ................................................................................................20 Net Unrealized Appreciation (NUA) .........................................................................................21 Summary ..................................................................................................................................24 Chapter 4: Computation of Substantially Equal Periodic Payments....................26

Minimum Method .....................................................................................................................27 Amortization Method................................................................................................................29 Annuity Method........................................................................................................................30

iv

Other Methods ..........................................................................................................................31 Chapter 5: Interest Rate Selection Risks ................................................................32

External Risk ............................................................................................................................32 Internal Risk .............................................................................................................................33 Internal Risk Management ........................................................................................................34 External Risk Management .......................................................................................................36 Conclusions ..............................................................................................................................37

Chapter 6: Life Expectancy Table Selection ..........................................................39 Approved Mortality Tables .......................................................................................................39

Table S: 80CNMST & 90CM....................................................................................................43 Chapter 7: Other Planning Issues...........................................................................45

How to Plan a SEPP..................................................................................................................45 Account Fracturing and Aggregation.........................................................................................45 Multiple SEPP Programs...........................................................................................................46 Important Dates ........................................................................................................................46 Stub Periods .............................................................................................................................47 Disbursement Frequencies and Locations..................................................................................48 Diminimus Issues......................................................................................................................48 Cost of Living Adjustments ......................................................................................................48 Recalculation in Conjunction with the Amortization or Annuity Methods .................................49 Death and Divorce ....................................................................................................................54 Simultaneous Processing of Other Exceptions...........................................................................55 Planning of the SEPP ................................................................................................................55 Tax Planning Issues for the Wealthy .........................................................................................56 Conclusions ..............................................................................................................................57

Chapter 8 Risk Analysis ..........................................................................................58

Risk Assessment .......................................................................................................................58 SEPP Program Design Risk Assessment ...................................................................................60 Individual Issues .......................................................................................................................61 Selecting a Professional ............................................................................................................61

Chapter 9 Administration .......................................................................................63

Record Keeping ........................................................................................................................63 Trustee Communications...........................................................................................................63 Tax Matters...............................................................................................................................63 State Taxation ...........................................................................................................................64

Appendices

Appendix A: Internal Revenue Code §72(t)...............................................................................65 Appendix B: IRC Reg. §1.72-17A(f) Disability.........................................................................69 Appendix C: Excerpt from Notice 89-25, 1989-1 C.B. 662 .......................................................70

v

Appendix D: Excerpt from IRS Publication 590........................................................................71 Appendix E: Table I from Publication 590: Single Life Expectancy ..........................................75 Appendix F: Table II from Publication 590: Joint Life and Last Survivor Expectancy...............76 Appendix G: UP-1984 Mortality Table .....................................................................................79 Appendix H: Sample Annuitant Divisor Table (Based on UP-1984)..........................................82 Appendix I: Sample 1099-R with Instructions ...........................................................................85 Appendix J: Sample Form 5329 with Instructions .....................................................................88

Glossary ...................................................................................................................94

vi

________________________________________________________________________________________________

Introduction At the dawn of the twenty-first century, many taxpayers have acquired substantial retirement assets in a variety of tax-deferred vehicles such as:

• Employer-sponsored plans, such as profit sharing plans, 401(k) plans, and the conversion or lump-sum distribution of older defined benefit plans.

• Personally initiated traditional individual retirement accounts, (IRAs) including those funded by annual contributions and rollovers.

• Simplified employee pension accounts (SEPs).

• Recently created Roth IRA accounts, through both regular annual contributions and conver-sions from traditional IRAs.

More and more frequently, taxpayers in their thirties, forties, and early fifties are voluntarily (or involuntarily) contemplating some form of early retirement and are perplexed as to how they can ac-cess their tax-deferred assets without undue tax burden, as required by the Internal Revenue Code (IRC).

As of October 2001, IRC §72(t) now provides 12 reasons or exceptions that qualify taxpayers to withdraw monies from their 401(k)s or IRAs (collectively called “deferred accounts1”) and avoid the 10% surtax. These exceptions generically fall into two types: transaction specific and process/multi-year. There are seven transaction specific exceptions, which will be discussed briefly later on the text. As an example, exception #10 is the ability to withdraw some monies while you are unemployed and need to make health insurance premium payments2. All seven of these transaction-specific exceptions are either severely limited in scope or are present for coordination purposes with other parts of the IRC. Conversely, the focus of this text is to closely examine the other five “process” or “multi-year” exceptions. They are:

• §72(t)(2)(A)(i) Age 59½

• §72(t)(2)(A)(ii) Death

• §72(t)(2)(A)(iii) Disability

• §72(t)(2)(A)(iv) Substantially equal periodic payments (“SEPPs”)

• §72(t)(2)(A)(v) Separation of service at age 55

These five code subsections deal very specifically with how a taxpayer (or his or her estate) may commence withdrawals from 401(k)s, IRAs, and other deferred accounts and avoid the application of the 10% surtax on such distributions.

Within this text, you will find issues and solutions that the general public, professional accountants, and lawyers seldom confront. Further, due to a general absence of concise and practical information on 1 Within this context, we include all savings plan type accounts, of a “defined contribution” nature, which are qualified un-der various sections of the IRC §§401-424. As always there is an exception, that being Roth IRAs, which are separately discussed in Chapter 3 due to their unique characteristics. 2 IRC §72(t)(2)(D)

vii

this subject, many professionals periodically end up in a “reinvent the wheel” mode. Every time the wheel gets reinvented, it tends to look a little different—sometimes better, sometimes worse—than the last wheel. This is somewhat dangerous in that, as we will discover later, making a mistake in theory or practice in this area can be extremely costly. We are faced with IRC sections that are becoming more popular in their application coupled with an absence of concise authority, further coupled with an ab-sence within the IRC to correct mistakes.

The scope of this text covers a variety of what we will collectively call deferred accounts. These include all forms of §401(a), §403(b), and §401(k) plans as well as virtually all IRA types, SEPs and SIMPLEs. Although §401(a) and §403(b) cover both defined-benefit and defined-contribution plans of various types, this text deals only with defined-contribution plans and accounts3. Frequently, we will use the specific term of “IRA” in an example. The reader should assume that the example applies equally to all types of deferred accounts unless there is a notation to the contrary.

The audience for this text is twofold: individual taxpayers and professionals. More and more individuals are contemplating or working through the financial issues of early re-tirement and need to withdraw from deferred accounts before the age of 59½. This text should answer almost all your questions in this circumstance. Professionals, primarily accountants, and lawyers cur-rently have no place to go when a client presents a situation requiring the implementation of §72(t)(2)(A) issues. Similarly, this text should solve almost all of those questions. However, as is need-less to say, it is always possible for a client to present a permutation of facts that will cause the answer to become unclear.

We have attempted to make this text as thorough as possible. Unfortunately, most of the authority written on §72(t)(2)(A) is spread over a variety of document types and locations—many of which would be unfamiliar to the average taxpayer. As a result, whenever possible, we have included the lit-eral text, tables, and other computations in the appendices.

Finally, we have made every attempt to explain the relevant theory and then bring that theory down to a practical level such that all readers should be comfortable with the basic concepts as well as the computational nuances. Ninety-nine percent of all published literature on “deferred accounts” is either persuasive in na-ture—meaning a focus on why you should have an IRA or why you should contribute to your em-ployer sponsored 401(k) plan—or it is focused on how you should manage these deferred assets during their accumulation years, e.g. should you buy stocks, bonds, mutual funds, etc. To the best of this au-thor’s knowledge, there is no other comprehensive literature or writing available on getting your money out; particularly on getting your money out early. That is what this text is all about. Those read-ers specifically interested in SEPP mathematics can jump forward to Chapters 4 through 7, all other readers are encouraged to read each chapter. The author is the first to admit that in-depth discussions on §72(t)(2)(A) are, on a scale of one to 10, a 13 in the dry reading department. As a result, we have occasionally inserted some humor, so eve-ryone can stay awake. On the other hand, humor can be dangerous, as someone can always be of-fended. We mean no offense, malice, or harm in this regard. 3Defined-benefit plan payments and annuities are ineligible for treatment under §72(t). Instead, these plan types are handled by IRC §72(q)(2) and are outside the scope of this text. However, as a general comment, almost all of the exceptions found in §72(t) are repeated in §72(q) to afford the same tax treatment for taxpayers making withdrawals from defined-benefit plans.

viii

________________________________________________________________________________________________

Second Edition This is the second edition of the text. The first was published September 2000. The second edition reflects all changes in the IRC during the intervening period. Further, the second edition has new or expanded sections on all 12 exceptions for avoidance of the 10% surtax, a new chapter covering Roth IRA penalty tax circumstances and the mechanics of implementing the “Net Unrealized Appreciation” method, and updates for private letter rulings issued in the last 16 months that periodically give us ad-ditional or new insights into IRS thinking on specific tactics. In addition, the author has attempted to incorporate many of the reader suggestions from the first edition, typically in the area of attempting to make some of the more technical passages a little easier to understand.

Unfortunately, the second edition grew in size to almost one hundred pages compared to the first edition size around seventy-five pages. Readers, don’t despair; the subject matter did not get 33% more difficult over the past year; rather, the breadth of material covered has been materially expanded as noted above. Further, our objective (in both editions) has always been to treat this subject exhaustively, primarily because readers have no place else to go. As a result, it is unlikely that anyone needs this whole text because of its dictionary or reference book like nature.

________________________________________________________________________________________________

Disclaimer In a way, it is unfortunate that books of this nature require disclaimers, but such is today’s legal environment. We have made every effort to be accurate and believe that to be the case as of October 2001. Unfortunately, the IRC and related documents of authority are ever-changing. As a result, in some small fashion, this text has a high probability of being out-of-date within several months to a year. As a result, we can offer no prospective promises regarding accuracy beyond the date of publica-tion. Recognizing that this does not really help the reader, we suggest that serious readers intent on im-plementing one or more of the strategies outlined in this text should confirm to themselves that they are in fact dealing with the most current set of facts and applicable law. The easiest way to do this is to hire the expertise of a tax accountant or a tax attorney who is conversant on these topics. ________________________________________________________________________________________________

Biographical Information William J. Stecker, (Bill), earned a bachelor’s degree in Finance and Applied Mathematics in 1971 and a Masters of Science in Accounting from DePaul University (Chicago, Illinois) in 1983. He is a Certified Public Accountant since 1980 and a member of American Institute of Certified Pubic Ac-countants as well as a member of both the Illinois and Colorado State Societies. Bill has spent 30 years in accounting, finance and related fields. Further, he has particular in-depth knowledge on this subject, having made multiple and successful private letter ruling requests and determination letter requests to the Internal Revenue Service.

ix

Bill and his wife, Mary, own The Marble Group, Ltd., an accounting, tax, and financial consulting group based in Colorado. Bill devotes most of his time to tax issues, specializing in retirement plan and participant issues—primarily §72 and §§401-424 of the IRC. You can send your inquires to: The Marble Group, Ltd., 30671 Bearcat Trail, Conifer, Colorado 80433 or w.stecker@att.net or themarblegroup@att.net.

1

________________________________________________________________________________________________ Chapter 1 Current Law In this chapter, we present the currently available resources that offer guidance on IRA withdraw-als that are not subject to penalties. These sources create something of a rule book. To help clarify these technical passages, we’ve added a brief explanation of each citation. Further, this chapter will cover all 12 exceptions and conclude with an introduction to Substantially Equal Periodic Payments, (SEPPs), exception #4.

Internal Revenue Code Internal Revenue Code (IRC) §72(t)(1) imposes a 10% surtax on all withdrawals from deferred ac-counts. Immediately following, however, IRC §72(t)(2) grants a variety of exceptions to §72(t)(1). It is important to remember that this is one of those sections of the IRC that was intentionally drafted in re-verse, e.g. a 10% surtax will always apply unless the taxpayer fully meets one of the exceptions identi-fied in §72(t)(2)(A) or elsewhere within §72(t). As a result, if the taxpayer does not fully satisfy an ex-ception, then the 10% surtax is re-applied by the Internal Revenue Service (IRS) without any need for further statutory action.

Exception #1 Age 59½ The best known exception is achieved by attaining the age of 59½4. This exception would appear, on its face, to be rather easy to apply, e.g. make a withdrawal after you are 59½ and you will not be subject to the 10% surtax. However, there are two very specific rules that apply:

• The IRS is required to make a literal age interpretation here because it is the IRC which spe-cifically says, “made on a day on or after the date on which the employee attains the age 59½.” So when do you have your 59th and ½ birthday? Exactly 183 days after your 59th birthday. As an example, if your birth date is April 26, 1946, your 59½ birthday is October 26, 2009. A withdrawal made on this date or later will be fine; a withdrawal made on October 25, 2009, or earlier will be subject to the 10% penalty5.

• As we will learn later, if you are making substantially equal periodic payment distributions sub-ject to IRC §72(t)(2)(A)(iv), exception #1 may be held in abeyance for some amount of time until all of the requirements of exception #4 are fully satisfied.

Exception #2 Death6 Death, although never really a desirable outcome, is usually a fairly straight-forward event; some one either is or is not deceased. This exception, however, is a good time to raise two topics: one, the IRC versus the IRS, and two, the concept of playing audit lottery. 4 IRC §72(t)(2)(A)(i). 5 Many other sections of the IRC are written in a manner that examines a taxpayer’s “highest attained age” within a particu-lar tax or calendar year. This is not one of those situations. Instead, the IRS is required to and does measure to the day. 6 IRC §72(t)(2)(A)(ii).

2

• By the time the reader reaches the end of this text, it will become apparent that there is often considerable latitude in the implementation of these exceptions but in other circumstances it will appear as if the rules switch around and become unduly harsh and restrictive. The reason for this lies in determining who or what is the prevailing authority on an issue. When the words are embodied in the IRC, then the IRS has no alternative but to take a very literal interpretation of the law—thus, this whole stickiness on when does some one truly turn age 59½.

Conversely, there are many circumstances where the IRC is vague or may embody language like, “based on regulations as issued by the Commissioner.” In this later case, decision-making authority is actually transferred from Congress to the IRS on the belief that the IRS is better at examining the details of a situation and is better qualified to issue appropriate regulations and opinions. In these later cases, we will find considerable additional flexibilities afforded to the taxpayer.

• Everyone would like to win the Powerball lottery, but this is about as likely as being struck by lightning three times on the same day. On the other hand, no one wants to win in the IRS lot-tery. Further, should your Social Security number magically turn up7, it is very important to know which lottery you are in. In the arena of auditable issues on §72(t) there are really two different lotteries: the congres-sional lottery and the IRS lottery. In same context as above, if you are audited and the central issue to be resolved lies in the interpretation of the IRC as drafted by Congress, YOU WILL LOSE! The IRS as well as tax courts and federal district courts will take a literal and restrictive construct of the IRC language. In short, there is zero maneuvering room. Conversely, if the central issue to be resolved lies in the interpretation of the IRC through the IRS’s issuance of regulations, rules, notices, etc., you may then stand a chance greater than zero of prevailing; that chance, however, is still relatively small. Sometimes the temptation to venture into “audit land and gray areas of the IRC” may appear to be a good business decision. This is not one of them. As we will discuss later in the text, the body of knowledge is not in the taxpayer’s favor, and in this particular section of tax law, there are virtually no corrective measures available to the taxpayer when the IRS prevails. There is no way to put back the improper distributions and amend your tax returns. Further, there isn’t even any negotiating room in the computation of the penalties and statutory interest due. In summary, §72(t) is not the place to push the envelope as the downside risks are substantial and almost always, non-negotiable.

The second exception, embodied in §72(t)(2)(A)(ii) says in part, “made to a beneficiary…on or af-ter the death of the employee.” If you are audited, simply make sure to take a copy of the signed death certificate with you to the examination.

Exception #3 Disability8 The full text of exception #3 says, “attributable to the employee’s being disabled within the mean-ing of subsection (m)(7).”

7 There is absolutely no evidence that taking early withdrawals from your deferred-asset account(s) in any way influences or increases your likelihood of an audit. However, should you be audited, there is a virtual 100% probability that your with-drawals will be closely examined to insure full compliance with the law. 8 IRC §72(t)(2)(A)(iii).

3

“Attributable” means that in order to use this exception, you, personally, must be physically and/or mentally disabled according to the following definition:

What does the preceding mean? First, with certainty, if you choose to use this exception and are subsequently audited, you will be required to present external proof and evidence that you are disabled. Second, the IRS has issued a fairly in-depth regulation on this subject. It says in part:

The above Code and regulation language imply several tests or hurdles that must both be met or exceeded in order to qualify for this exception. They are:

• Effectively, the disability must be total as opposed to partial. As an example, some one may be disabled in all outward appearance; however, if that person were capable of (irrespective of whether he does or not) returning to work on a part-time basis; then that person would not be considered disabled under this definition.

• The disability is always measured in context to the individual’s pre-disability occupation. As an example, two brothers are in a vehicular accident resulting in both brothers becoming para-plegic; however, both subsequently fully recover but for the use of their legs. One brother is a professional football player and the other is a surgeon. The former is disabled; the later is not.

The IRS will consider any and all external evidence you produce but is not necessarily bound or obligated to honor that evidence in making its determination. Such evidence might include doctor statements, employer statements, proof that you are already collecting private disability insurance, and Social Security disability proceeds—Social Security being the most persuasive evidence. Further, be-ing disabled has been classified as a “facts and circumstances” test by the IRS; accordingly, they will not rule in advance on this issue for individual taxpayers. As result, this author believes that using ex-ception #3 is a fairly dangerous strategy unless the disability is beyond a shadow of a doubt. With any lesser disability, the taxpayer is well advised to seek the professional opinion of an attorney who spe-cializes in disability matters.

“In determining whether an individual’s impairment makes him unable to engage in any substantial gainful activity [emphasis added], primary consideration shall be given to the nature and severity of his impairment. Consideration shall also be given to other factors such as the individual’s education, training, and work experi-ence. The substantial gainful activity to which section (m)(7) refers is the activity, or comparable activity, in which the individual customarily engaged prior to the arising of the disability….”

IRC Reg. §1.72-17A(f)

“…an individual shall be considered to be disabled if he is unable to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or to be of long-continued or indefinite duration. An individual shall not be considered to be disabled unless he furnishes proof of the existence thereof in such form and manner as the Secretary may require.”

IRC §72(m)(7)

4

Exception #5 Separation of Service at Age 55 Wait a minute, what happened to exception #4? Number 4—substantially equal periodic pay-

ments—is by far the most difficult of the process/multi-year exceptions to understand and is the most difficult of all of the exceptions to implement correctly. As a result, we are going to skip #4 for the time being and finish the discussion of the remaining exceptions first.

In almost all cases in this text, we treat “deferred-plan assets” (e.g. assets accumulated within plans qualified under §401(a), §401(k), §403(b), and IRAs) and IRAs as being the same and accordingly re-ceive the same treatment under §72(t)(2)(A). IRC §72(t)(2)(A)(v) is an exception to the general rule. Subsection (v) provides an ability for a separated employee to commence withdrawals provided he is 55 or older when separated from the employer who is the plan sponsor from which the now terminated employee wishes to make withdrawals. Unfortunately, §72(3)(A) says in part that §72(t)(2)(A)(v) “does not apply to distributions from an individual retirement plan.9” This simple sentence effectively forces a split between plan assets and IRA assets and makes subsection (v) available only to plan assets and disallows the same treatment for IRAs. Notice 87-13 provides a further amplification of how to interpret §72(t)(2)(A)(v) by saying:

Finally some reasonably clear language. As we will learn when we get into the approved methods and mathematics of SEPPs, SEPPs by their nature are restrictive. Here, we can skip all of the SEPP rules if the taxpayer can comply with the rules enumerated in Notice 87-13. What are those rules? 1. The taxpayer must terminate employment from his or her employer in the same tax year as he or

she attains the age 55 or greater. This is not the same as being 55 when you quit. Instead, one need only reach age 55 anytime in the same year, even if attaining age 55 occurs six months after termi-nation. Thus a theoretical minimum age exists of 54 years and two days. If your birthday is De-cember 31, you could quit on January 2 of the year you turn 55.

2. The assets withdrawn must be withdrawn from the plan sponsored by the employer from which the employee has just terminated. This is critical. Further, there are two subsidiary issues of equal im-portance:

• If the taxpayer terminates employment with the employer/plan sponsor and immediately rolls the plan assets into an IRA; the IRA assets, though they did originate from the qualified plan become ineligible for treatment under §72(t)(2)(A)(v). The assets absolutely must remain in the plan. Thus, as a caution, taxpayers should resist the sales pitches from various financial advo-cates that ex-employees should quickly roll over their plan assets into IRAs in order to take ad-vantage of all the investment vehicles that are available through an IRA and might not have been available inside the plan. If you are 35, go ahead; if you are 54½, this can be a disaster.

9 Why this distinction is made between plan assets and IRA (definitionally not a qualified plan) assets remains a mystery.

“Section 72(t) (as added by TRA '86) applies an additional tax equal to 10 percent of the portion of any ‘early distribution’ from a qualified retirement plan...that is includible in the taxpayer’s gross income.... A distribution to an employee from a qualified plan will be treated as within section 72(t)(2)(A)(v) if (i) it is made after the em-ployee has separated from service for the employer maintaining the plan and (ii) such separation from service occurred during or after the calendar year in which the employee attained age 55.”

IRC Notice 87-13

5

• Working for a single employer for thirty or more years is pretty much an event of the past. As a result, it is not uncommon for a taxpayer to have three, four, or even five 401(k) plans and/or related conduit IRA accounts10. Further, it tends to be the oldest accounts that are the biggest because they have had the longest time to grow. Nonetheless, Notice 87-13 tells us that we can make withdrawals only from the plan account from the employer through which the employee just separated. How do we get at the plan assets from the prior employers? The easiest solution is to perform a series of rollovers. Roll the assets from either prior employer plans or conduit IRAs forward into your current employer’s plan. Further, these rollovers need to made before you terminate, as rollovers are often prohibited for terminated plan participants. On its face, this does appear a little illogical. How can it make a material difference if all the money is in the current plan versus spread out in four different plans. Unfortunately, this is one of those situations where form does govern substance.

Thus, taxpayers who are 55 (or will be 55 during the current tax year) and are separating from ser-vice from their employers11 should pause and think very carefully about this issue. There is no clear-cut obvious or winning decision here, as there are pros and cons to each decision path. The biggest ad-vantage is that the taxpayer need not pay any attention to SEPPs and instead can withdraw any amount at any time12. The biggest disadvantage is that most 401(k) plans limit investment choice (often to mu-tual funds only), which the taxpayer may not like. Any taxpayer in the 54½ to 59½ age window leav-ing employment should seek professional help before automatically rolling over 401(k) assets into an IRA.

Exceptions #6 Through #12 These exceptions are referred to as the transaction exceptions in that they tend to be very limited in scope and are often one-time events.

• #6 §404(k) Stock Dividends13. This will likely occur whenever you are member of an older §401(a) plan in which the plan sponsor/employer contributes employer stock to the plan. Fur-ther, the plan was drafted in a manner that requires that when the corporation declares a divi-dend, the dividend must be paid on the securities held in the plan. Strangely enough, the actual dividend dollars cannot be paid into the plan itself but must instead be paid directly to plan par-ticipants. In short, almost all readers need not concern themselves about this exception. If you are subject to it, your plan administrator will let you know how to declare the dividends re-ceived as income and specifically how to avoid the 10% surtax.

10 We use the term “conduit IRA” account here to mean a temporary IRA account housing the assets of a prior employer’s qualified plan, usually waiting to become eligible to participate in a new employer’s 401(k) plan. 11 In today’s economic environment, many taxpayers are being offered “early retirement” packages from their employers. Often, an employer will be willing to structure severance pay for an early retiree such that the periodic severance pay and therefore employment extends until the first weeks of January of the tax year in which the taxpayer will turn age 55 thus making this exception a viable alternative. 12 This is all predicated on being a participant in a qualified plan that provides both for periodic payments to participants and a reasonable and cooperative plan administrator. Each taxpayer facing this situation should check with his or her re-spective plan administrator as soon as possible to have a detailed discussion regarding plan provisions, particularly the abil-ity to support and process periodic payments. 13 IRC §72(t)(2)(A)(vi).

6

• #7 Tax Levies14. Should you be unfortunate enough to be subjected to a federal tax levy15, Congress has graciously determined that the 10% surtax will not be due. However, there are several catches: one, regular federal income tax will still be due upon the distribution of the IRA; and two, the distribution of the IRA must be pursuant to a court ordered levy and the dis-tribution must be made directly to the IRS. Conversely, if the IRS sends you a deficiency notice or otherwise obtains a judgment against you and you voluntarily cash in an IRA to satisfy the deficiency or judgment, this exception will not apply.

• #8 Medical Expenses16. You may make distributions from your IRA up to the amount of your qualifying medical expenses17 and escape payment of the 10% penalty remembering that regu-lar federal income tax will still be due on the amount distributed. However, the amount one can withdraw using this exception is limited to those qualifying medical expenses that are in excess of 7½% of your adjusted gross income.

• #9 Qualified Domestic Relations Orders18, (QDROs). A QDRO is a court order signed by a judge typically as a result of a divorce proceeding. With respect to your IRA, the judge can conceivably sign two different types of QDROs. The judge may sign an order requiring the di-vision of your IRA into two IRAs; one for you, and one for your spouse, where the character of the newly created IRA remains the same simply changing the name on the new account. This type of IRA division is governed by IRC §414(p) and is a nontaxable event. Or, the judge may order a payment stream (thus creating a taxable event) from your qualified plan19 to one or more alternate payees—typically, an ex-spouse and/or children. In this case a taxable event has occurred requiring that income tax be paid on the amounts distributed each year. Fortunately, the 10% surtax is excused, irrespective of your age.

• #10 Payment of Health Insurance Premiums While Unemployed20. If you have been or were unemployed for 12 or more weeks, you can make withdrawals from your IRA up to the amount of your health insurance premiums for the period during which you were unemployed.

• #11 Higher Education21. Generally speaking, distributions from your IRA can be made to pay for post-secondary qualifying educational expenses for yourself, your spouse, and your chil-dren. Seeking professional assistance is advised as the “qualifying” expenses part of this rule does get a little tricky.

14 IRC §72(t)(2)(A)(vii). 15 This is tantamount to the IRS performing asset seizure, where it will typically take your residence, your cars, your jew-elry, and, by the way, your IRA as well—all pursuant to court order. 16 IRC §72(t)(2)(B). 17 See IRC §213; typically the same medical expenses you would place on Schedule A -- Itemized Deductions. 18 IRC §72(t)(2)(C). 19 Please note the change in language here to “qualified plan.” IRAs are not eligible for this exception. 20 IRC §72(t)(2)(D). 21 IRC §72(t)(2)(E).

7

• #12 First Home Purchases22. Generally, you may withdraw up to $10,000, per lifetime, to purchase a principal residence—provided that you have not owned a home in the last 24 months. The detail rules here can get very complex.

A word of caution. Exception rules #6 through #12 all vary in complexity, each in its own right. Above is a brief synopsis simply to let taxpayers know that they exist and may be used when the proper conditions present themselves. Exceptions can generally be used independent of another; how-ever, there is always one right way and usually a dozen wrong ways to implement an exception. Al-though some of these exceptions appear to be fairly generous, particularly for distressed circumstances, it is advisable to seek professional assistance on any and all of these because an exception, incorrectly implemented, will be disallowed and the 10% surtax plus interest will be re-imposed. Further, as with all of the exceptions in §72(t), there are no corrective measures that can be taken once an exception has been improperly implemented.

Exception #4 Substantially Equal Periodic Payments IRC §72(t)(2)(A)(iv) introduces us to a lesser known and the least understood exception—the con-cept of a “substantially equal periodic payment” (SEPP). Of particular importance is what the IRC says about this concept:

Note: A complete excerpt of IRC §72(t) is included in Appendix A. We can learn a lot about SEPPs from just this one sentence. It is actually easiest to take the term “substantially equal periodic payments” and deal with each word in reverse order. PAYMENTS in this case mean a dollar23 withdrawal from a deferred account. Further, because of the plural, we can infer that there will need to be more than one payment. PERIODIC implies that each withdrawal will occur over some set of defined intervals. Immedi-ately following we find the language “not less frequently than annually.” As a result, we know the out-side limit with certainty—a minimum of once a year. However, there is no guidance about more fre-quently than annually. As we will learn later, monthly and quarterly withdrawals are fine as well. EQUAL implies that the withdrawals from the deferred account will be of equal dollar amounts.

SUBSTANTIALLY is potentially the most interesting word in the sentence. It becomes a modifier to the word “equal.” If Congress had intended for these payments to be literally equal, it would not have used the adjective “substantially.” As a result, we now have “sort of” or “almost” equal payments because we do not yet know how to interpret the word “substantially.”

22 IRC §72(t)(2)(F). 23Actually, withdrawals can be “in kind” as well as cash. All of our examples will focus on cash withdrawals or disburse-ments from one or more deferred accounts.

“…part of a series of substantially equal periodic payments (not less frequently than annually) made for the life (or life expectancy) of the employee or the joint lives (or joint life expectancies) of such employee and his desig-nated beneficiary.”

IRC §72(t)(2)(A)(iv)

8

The remainder of the definition is devoted to time span; namely the life or life expectancy of the employee24 or employee and beneficiary. We can interpret this to mean that there is an outside parame-ter at which time payments will stop. Next we need to turn our attention to IRC §72(t)(4)(A)(ii)25, which gives us some rule modifiers as follows:

Essentially, this says that, barring death or disability, if the taxpayer modifies or discontinues these SEPPs before the passage of at least five years and attaining the age of 59½, then the 10% surtax plus interest will be reimposed. As we will learn later, the word “modified” will become critical in our thinking and strategy. As a result, we can now develop a preliminary working definition of SEPPs as follows: A series of payments or withdrawals from deferred accounts that can be either exactly equal or potentially some-what dissimilar, that occur annually or more frequently, and continue for a minimum time period of five years or until the taxpayer reaches age 59½ , whichever is greater. So, if we set up a payment stream that meets all these criteria, we should be able to implement that payment stream and avoid the 10% surtax. The only other modifiers or rules to be found in the IRC are some rather obvious ones:

1. The 10% penalty, if applied, applies only to that portion of the distribution that is includible in gross income.

2. An employee must have separated from service in order to commence SEPPs that include dis-bursements from an employer plan26.

Virtually all of the remainder of this text is devoted to avoidance of the 10% penalty tax. Unfortu-nately, regular income tax (both federal and often state) is always due. After all, both contributions by employees to 401(k) plans and deductible IRAs and employer contributions to various plans as well as all of the earnings accrued through the intervening years have never been taxed. Thus, we tend to call all of these accounts and assets as “deferred asset accounts,” not “tax-free accounts.” Taxes have been 24 Frequently, the IRC uses the term “employee” in the context of one being employed and therefore a participant in a quali-fied plan. The reader can also conceptually substitute the word “owner” in the context of an IRA. 25 IRC §72(t)(4)(A) is pretty clear in the circumstance when a taxpayer, aged 50, commences SEPPs and modifies them before age 59½; he will owe back taxes and interest on 100% of the distributions. What about a taxpayer, aged 56½ who makes SEPP distributions to 59½ (3 years) and then modifies those SEPPs after age 59½? Our taxpayer is more than 59½ but has failed the five-year test. In this case, the surtax and interest will be imposed only on the three early year’s worth of SEPPs. See Pub 575, page 29, 2000 edition. 26 In order to remain “qualified” virtually all plans have “separation of service” language in them to prevent withdrawals from the plan while still employed. Some plans, however, will permit “in-service hardship withdrawals” under certain spe-cific conditions.

“ [If]...the series of payments under such paragraph are subsequently modified (other than by reason of death or disability)… before the close of the 5-year period beginning with the date of the first payment and after the em-ployee attains age 59½, or… before the employee attains age 59½, [then]...the taxpayer’s tax for the 1st taxable year in which such modification occurs shall be increased by an amount... equal to the tax which...would have been imposed plus interest for the deferral period.”

IRC §72(t)(4)(A)

9

deferred on these accounts during their accumulation or building years. Now that the taxpayer wants to commence withdrawals, the tax-deferral period ends and Congress, through the arm of the Internal Revenue Service, is standing at the door, ready to collect. Thus, we are willing to (although begrudg-ingly) to pay regular federal income tax on those distributions; however, we are unwilling to pay a 10% surtax or penalty just because we are withdrawing assets before the age of 59½.

Internal Revenue Service Publications Next on our document list is Notice 89-2527, which gives us one page of information as to what the IRS believes constitutes a “series of substantially equal periodic payments.” Finally, we start to get down to some specifics with this notice in that it provides us with three detailed methods of how SEPPs may be computed. These three methods are called the minimum, amortization, and annuity methods. Rather than get into the specifics of each here, we later devote a full chapter to these methods and the computational specifics of each. We can also take a look at Publication 59028, the IRS’s seminal text on IRAs. The relevant portion of Publication 590 is included in Appendix D. Of particular relevance is the section that is titled “An-nuities” commencing in the middle of the second page of Appendix D. Unfortunately, this reference in Publication 590 sends us right back to Notice 89-25 for detailed guidance29, which is discussed above.

Tax Court Cases Where else can we look for authority on this issue? Our next stop is the courts. Fortunately, here we find some new information in the case of Arnold v. Commissioner30. The real issue decided in this case was a precise definition of the five-year rule embodied in §72(t)(4)(A)(ii). The Arnolds contended that “five years” meant “five tax years;” the Commissioner contended that “five years” meant a literal interpretation of the calendar, e.g. if your first withdrawal was on 5/1/92, then five years later is 5/1/97 or 1826 days later. The Commissioner prevailed. There are other tax court cases that cite or make ref-erence to §72(t) but are not relevant in refining our definition of SEPPs.

Private Letter Rulings Our last source of authority lies in the private letter rulings (PLRs) issued by the IRS; of which there are approximately 75. We face a bit of a dilemma, however, when delving into PLRs. Every PLR issued by the IRS starts with “This document may not be used or cited as precedent. Section 6110(j)(3) of the Internal Revenue Code.” There is actually good reason for this language. PLRs are a response of law (or at least the IRS’s position of the law to which they are essentially bound) to a particular fact set from a particular taxpayer. Rarely are circumstances identical, and it would be much too easy for tax- 27 1989-1 Cumulative Bulletin 662. The relevant portion is contained in full text in Appendix C. 28 Every individual and practitioner should keep a copy of Publication 590 handy. It is a comprehensive text on IRAs and other retirement issues. 29 As we will ultimately learn, the IRS thinks “detailed guidance” is the logical equivalent of providing some one a map of the solar system in order to help locate North Platte, Nebraska. 30 111 TC 250; 1998 U. S. Tax Ct. Further, what appears to be of significance is the IRS’s very stringent interpretation of the IRC; versus its somewhat relaxed interpretations of IRS-originated documents on other §72(t) issues.

10

payer number two to use taxpayer number one’s PLR with a slightly modified fact set. Unfortunately, it can be that slight modification in the fact set that nullifies or reverses the essence of the PLR.

In addition to the obvious benefit of the submitting taxpayer, PLRs are published and are therefore the IRS’s method of providing other taxpayers and practitioners a “peek in the box” as to how the IRS is thinking on an issue, without the IRS being bound by its thinking (other than to the submitting tax-payer). Thus, at one level, PLRs are worthless to us. We can never pick a PLR or two and present them as a defense to the IRS in or out of a court of law31. At another level, PLRs become extremely impor-tant. It is in PLRs that we find almost all of the meat we need to make effective tactical decisions on SEPPs. Often we can find the same issue repeating itself across multiple PLRs resulting in the same decision by the IRS. This, at least, can provide us with some comfort in our analysis.

Further, as we will see in the risk assessment chapter, there are certain basic interpretations on SEPPs that are very safe, and PLRs simply provide an affirmation of our thinking. In other areas there may be one or at most two PLRs indicating that possibilities exist in this area, but the risks are higher. Also, there are several negative PLRs32 that represent very high risk levels. Moreover, it is always pos-sible to frame the question in a manner or with a fact set for which no ruling exists. When this occurs we at least know that we are treading in virgin country, which usually represents the highest of risk environments. Based on all that we know, from the IRC down to the PLRs, we will develop a risk as-sessment profile of SEPP programs and classify those SEPP programs as either below or above a sub-jective threshold where making a new private letter ruling request may be advisable.

31 This is not entirely true. While an individual PLR cannot be admitted into evidence as a defense, a group of PLRs—treated as a group—can be admitted into evidence where the purpose of doing so is to abstract the theory espoused and its consistency throughout the sum of the PLRs while ignoring or discarding the transactional details, facts, and circumstances. 32 95% of all PLRs issued in this arena are “positive” meaning in favor of the taxpayer. Negative PLRs are unusual in that a submitting taxpayer is afforded the opportunity to withdraw or modify the PLR submission instead of receiving a negative reply. Then again, some taxpayers are simply stubborn.

11

________________________________________________________________________________________________

Chapter 2 Common SEPP Concepts Before we delve into the intricacies of individual SEPP programs, we should spend some time on the basics and concepts common to all SEPP programs. Central to all deferred accounts is the concept of timing. In exchange for the tax deferral granted in earlier years, each of us implicitly agreed to live by Congress’s implied definition of the word “retirement.” It does not matter if we like or dislike the congressional definition or are financially able to retire early; we signed on the first day we opened an IRA or made a contribution to a 401(k) plan.

Unearned Ordinary Income Congress has effectively legislated that withdrawals from deferred accounts are the logical equiva-lent of retirement33 and that retirement is essentially a time period or periods during which a taxpayer’s unearned income replaces earned income. Earned income is easy to understand; it is those monies re-ceived, typically from an employer, for your personal services, e.g. wages, salary, bonuses, and the like. Further, each taxpayer was taxed on this earned income when earned except for the contributions to deferred account savings vehicles, like 401(k) plans and deductible IRAs. By law, all the contribu-tions you and maybe your employer made to these deferred accounts as well as all earnings and in-vestment asset appreciation are reclassified as unearned income. As a result, some years later, the con-tents of your deferred account(s) are as yet untaxed34. Neither the manner or original source of the con-tributions nor the manner in which asset appreciation occurred is of significance. All withdrawals from all deferred accounts are treated as unearned ordinary income35 and are accordingly taxed at the then present graduated tax rates for ordinary income.

Age Windows In concert with indirectly defining the word “retirement,” Congress has created three time win-dows:

• Before age 59½

• Between ages 59½ and 70½

• After age 70½

33 This is not entirely true in that §72(t)(2)(A), (B), (C), and (D), as discussed in Chapter 1 contain a variety of “bad” and “good” specific/emergency situations, e.g. death, disability, education, first home purchase, etc. where the 10% surtax is waived; however, our main thrust of §72(t) is to deal with retirement situations. 34 The term “untaxed” in this context is referring to federal income tax. Conversely, if you examine your pay stub or year-end W-2 carefully, you will see that your contributions to your 401(k) plan were not federally taxed, but they were taxed for FICA/Medicare. Because contributions were taxed for FICA/Medicare on the way in, they are not taxed again on the way out. 35 There are two potential exceptions here: net unrealized appreciation (NUA) on employer securities discussed later and taxpayer basis in a deferred account. These topics are discussed later in the text.

12

Before age 59½ is called “early retirement.” During this period, Congress intentionally makes it difficult to make deferred-account withdrawals. It matters not that you were professionally successful or started a personal savings program early in life. Congress has deemed that individuals leaving the earned income workforce and substituting unearned income before age 59½ are early retirees. Con-gress further says, by implication, that this should not happen. Therefore it imposes extra taxes (namely the 10% surtax) on those individuals who make these early withdrawals. Fortunately, Con-gress did give us some narrow windows of opportunity. That is what this text is all about. Stage two, from ages 59½ to age 70½ is typically called “normal retirement.” During this period taxpayers have virtually unlimited freedom in deferred account withdrawals36. The only requirement is that the taxpayer pay ordinary income taxes on those withdrawals because the withdrawal dollars have not been previously taxed. Stage three really doesn’t have a formal name, but it starts at age 70½. At this point Congress re-verses course and, instead of making it difficult to make withdrawals, Congress mandates withdraw-als37, essentially forcing you to make withdrawals even though you might not want to. These are called minimum distribution requirements, and not meeting them is painful to the tune of a 50% excess ac-cumulations tax38.

The entire focus of this text is on early retirement withdrawals and the use of the few windows of opportunity Congress left open. Nonetheless, it is equally important to remember that your deferred accounts will live with you through each of these three different stages of tax treatment.

The Account Concept A taxpayer (or taxpayer and spouse) is almost always treated as a single taxable entity. Thus, if you think about IRS Form 1040, it starts to make a lot of sense. Through the use of a blizzard of subsidiary forms (e.g. one for interest and dividends, another for capital gains, and another for farm income) all taxpayer income eventually flows upward and lands somewhere on Form 1040. It is on this form that the whole taxpayer’s income situation is revealed and a tax is assessed. IRC §72(t) and SEPPs work in the exact opposite manner. Instead of treating the taxpayer as a sin-gle taxable entity, the taxpayer is fragmented into pieces. As we will later learn, this will often work to our advantage. SEPPs are structured, evaluated, and used on an account-by-account basis; not on an aggregate “taxpayer” basis.

For example: It is quite conceivable that a taxpayer might have three or six or nine separate de-ferred accounts. To commence SEPPs, the taxpayer will have two fundamental choices:

1. Select one of those nine deferred accounts and commence the withdrawal of SEPPs from that, and only that, account. The other eight accounts are held on the sidelines for future use at future dates.

Or 2. Select more than one of nine—up to selecting all nine—and define a SEPP universe; let’s say ac-

counts A, B, and C. SEPPs are then computed on the sum total value of all three accounts, and SEPP withdrawals can be made in any proportion from any of the three accounts defined in the

36 Previously, §4980A(c) imposed an excess distribution tax equal to 15% of the amount distributed in excess of an indexed value approximating $150,000. Fortunately, this section of the IRC was repealed effective 12/31/96. 37 See IRC §401(a)(9)(C). 38 See IRC §4974(a).

13

universe. Once a universe is defined, however, and the first SEPP withdrawal has occurred, the contents or membership of the universe is frozen and cannot change. For example, after a with-drawal has been made, deferred account X cannot be added into the mix of accounts A, B, and C.

This concept of fracturing the taxpayer’s deferred accounts applies only before the taxpayer reaches age 59½ and will work to our advantage in ultimately designing multiple SEPP programs that can run concurrently. Between the ages of 59½ and 70½, this concept becomes irrelevant. When the taxpayer reaches age 70½, Congress reverses course and requires that all deferred accounts be aggre-gated together for purposes of measurement and computation of minimum mandatory distributions.

Individual Ownership Deferred accounts are always owned individually39. There is no such thing as joint ownership of a deferred account. Thus, in addition to the account concept discussed above, married taxpayers are split apart when considering SEPPs. This is actually good news. As a result, one spouse can commence SEPPs while the other spouse does not. Further, the application of the SEPP rules, such as duration and the age 59½ rule are applied individually to each separate SEPP program launched by each spouse. For a working couple now contemplating early retirement, the situation starts to look like a matrix with the horizontal axis being “his and hers” and the vertical axis becoming the different deferred ac-counts owned by each individual. The lowest common denominator is any intersection point in this matrix as represented by a person/account permutation. This is the beginning level, the “atom” if you will, of SEPPs. SEPPs can aggregate vertically, as we discussed above on accounts. SEPPs can never aggregate horizontally. As an example, John and Cathy have both had working careers and have accu-mulated a variety of deferred account types from multiple employers:

John Cathy Deferred Account Type Amount Acct. # Amount Acct. #

401(k) $400,000 1 $65,000 6 Lump-sum distribution of an old “DB” plan $75,000 2 $25,000 7 Self-funded IRA $300,000 3 $150,000 8 SEP IRA $100,000 4 $200,000 9 Rollover IRA from prior employer’s 401(k) $200,000 5 $85,000 10 Total $1,075,000 $525,000

Between the two of them, 10 different accounts exist totaling $1.6 million. Valid combinations of these accounts to create a “SEPP Universe” would be any or all of John’s accounts (1, 2, 3, 4, and 5) or any or all of Cathy’s accounts (6, 7, 8, 9, and 10). Combining any of John’s accounts with any of Cathy’s accounts would be invalid.

39 It is easy to forget that the “I” in IRA stands for “individual.”

14

Relationship to Earned Income During your working career, you more than likely received earned income in the form of a pay-check from your employer. Conversely, deferred-account withdrawals are classified as unearned40 in-come. These two types of income are unrelated in that neither prohibits the creation or receipt of the other. Thus, it is perfectly okay to begin receiving distributions from your IRA in the form of a SEPP and at the same time continue to be gainfully employed (or self-employed) as you so choose.

Reversibility and Errors In many cases the IRC provides methods for a taxpayer to reverse course, sometimes years after an initial tax decision has been made. This is NOT one of those cases. Within our general context of dis-cussion, SEPPs are never reversible. However, if we broaden our scope momentarily, there are two circumstances where SEPPs can be reversed. They are:

Annual rollover—A taxpayer can always make a withdrawal from a deferred account and replace those withdrawn funds within 60 calendar days. This is then classified as a rollover, and each tax-payer is allowed one per year per account. It is not our intent to make use of this feature, but can potentially provide a limited ability to correct mistakes41.

Surtax—A taxpayer can always stop the SEPP withdrawals at any time and suffer the wrath of §72(t)(4); namely the application of the 10% surtax plus statutory interest. Again, it is not our in-tent to take advantage of this feature; further, it is pretty hard to conceive of this as any kind of ad-vantage at all. However, in subsequent chapters we will find a unique circumstance or two when it is actually in the taxpayer’s best interest to invoke §72(t)(4) early in some “pay me now or pay me even more later” scenarios.

However, the issues just discussed above are really exceptions. The point here is that each taxpayer should view the commencement of a SEPP program as irreversible. Further, the IRC does not differen-tiate between a change in taxpayer circumstance42 and errors. Errors can be mistakes in theory as well as practice. A taxpayer may believe he or she has properly interpreted how to apply one of the ap-proved methods. The taxpayer may be wrong resulting in a “theory error.” Or, a taxpayer may have theory down pat and properly interpreted but may make a date or math computational error. This then becomes a “practice error.” In any event, the IRC treats changes in taxpayer desire or circumstance, theory errors, and practice errors equally. All three events are unfortunately treated as “modifications” resulting in the application of §72(t)(4), the 10% surtax plus interest.

40 There is really no prejudice intended in the words “unearned” versus “earned.” If you have it, of course you earned it; it is simply a matter of which asset you employed: your time versus your other financial assets. Thus, these terms are really just titles for us to keep clear as to how current period or future period taxation will take place. Further, it is a common oc-currence for a taxpayer who commences SEPPs to wish also to make an IRA contribution. This is not permitted. Instead a taxpayer must have earned income from some other source and then make an IRA contribution based on that earned in-come. 41 There is also a strategy wherein a taxpayer creates seven different IRA accounts of approximately equal size and makes overlapping withdrawals starting on the 1st day, 54th day, 108th day, etc.; e.g. make a withdrawal on day 1 of $10,000 from IRA A; make a second withdrawal from IRA B of $10,000 on 54th day placing that money back in IRA A to satisfy the 60 day rollover rule. In this manner, a taxpayer can essentially lend oneself approximately 1/7th of the sum total IRAs value. This strategy does work, but this author thinks that it is a bit too much work involved compared to the gain received. 42 You might win the lotto after two years of SEPPs and desire to suspend those SEPPs or modify them to a smaller amount for the foreseeable future.

15

What should we take away from this? There are probably several points. First, a taxpayer should think long and hard before commencing SEPPs to insure that the SEPP program fits correctly with both upside and potential downside changes in personal finances. Further, this implies a great deal of thought as well as preplanning in advance of SEPP commencement. Second, a taxpayer who is the least bit hesitant should purchase some insurance against committing an error. The easiest way to do this is to hire a competent tax accountant and/or tax attorney43 who is insured and who is capable of performing or at least reviewing the taxpayer’s assumptions and computations. Thus, in the event an error occurs, the taxpayer can rely upon the professional’s errors and omissions insurance coverage.

Basis in Account So far, our entire conversation about SEPPs has assumed that the entire deferred-account balance is taxable as unearned ordinary income. This is not always the case. The most common exception is the individual who has made a number of nondeductible IRA contributions to the deferred account in prior years. Thus, we now have an account that has a mixture of taxed and untaxed monies in it. In this situation, the taxpayer has “basis” in the account, which are those contribution dollars that have al-ready been taxed.

For example: Assume a taxpayer wishes to commence SEPPs on an IRA with a current total valuation of $100,000, of which $20,000 was originally from after-tax contributions. The basis in this account is $20,000. Further, let’s assume that the SEPP amount for the first year is computed to be $5,000. Only $4,000 is includible in the gross income of the taxpayer; $1,000 is excluded as a ratable return of basis in the account.

This return of basis issue can get considerably more complex in year two when accounting for ad-ditional account appreciation or when a taxpayer selects a multiple-account SEPP universe. Since IRS Publication 590 does an excellent job of covering most of these situations, we will not repeat them here. However, the basic concept continues into future years—that being a mathematical computation to arrive at a percentage or dollar amount representing a ratable return of basis, e.g. prior-year previ-ously taxed contributions that are not re-taxed when distributed.

Penalty Computations Almost all of this text is explicitly devoted to avoiding penalties. Nonetheless, a common charac-teristic of all SEPP programs is the danger that, for some reason, penalties will be enforced. IRC §72(t)(4) says in part that if a taxpayer modifies the SEPPs:

What does this mean? This is essentially a “look back” tax that says that if a modification occurs, then an additional tax is going to be immediately due in and for the same tax year in which the modifi-cation occurs. 43 We will discuss how to hire competent professional assistance in the risk assessment chapter.

“The taxpayer’s tax for the 1st taxable year in which such modification occurs shall be increased by an amount...equal to the tax which...would have been imposed, plus interest for the deferral period .”

IRC §72(t)(4)

16

This additional tax is computed by going back to the first year in which the SEPPs started and for that year and every intervening year recomputing the tax due for each year applying the 10% surtax. Thus, we wind up with an additional tax due by year since the commencement of SEPPs. Secondly, for each year in which additional tax is due and it is not the current year, compute the statutory interest due for all intervening years as a penalty for not having originally paid the tax when it was due.

For example: A taxpayer, aged 52, commenced SEPPs of $10,000 per year and continued same for five years, the year in which he reaches age 57. In year six, when the taxpayer turns 58, he with-draws $20,000. This will clearly be classified as a modification that violates the greater of five years or attaining age 59½ rule. Given that he has already completed six years, he would owe an approximate penalty of $8,19944. Referring to the following table, we can see that the aggregate penalty increases at a geometric pace the further one progresses into a SEPP program.

Notice the interest in years one through three is pretty meaningless such that a modification in year three would cost the taxpayer approximately 32% of an annual withdrawal amount. However, by year seven, the penalty is not 70% of an annual SEPP amount, but 87%. This percentage leaps to 140% by year 10. Also, please note that the 10% surtax plus interest is not only due on just the modified withdrawal; essentially, the entire payment stream from inception to completion is disallowed; thus the 10% pen-alty plus interest is assessed on all dollars withdrawn.

As a result, if a taxpayer is ever tempted, typically by a material change in personal circumstance or finance, to intentionally modify the SEPPs, he should clearly do so early in the SEPP program as opposed to later.

Approximate Computation Of §72(t)(4) Surtax and Interest Year Beginning

Balance Penalty

Assessed Statutory

Interest Charged Ending

Balance Due 1 $0 $1,000 $0 $1,000 2 $1,000 $1,000 $73 $2,073 3 $2,073 $1,000 $150 $3,223 4 $3,223 $1,000 $234 $4,456 5 $4,456 $1,000 $323 $5,779 6 $5,779 $1,000 $419 $7,199 7 $7,199 $1,000 $522 $8,720 8 $8,720 $1,000 $632 $10,353 9 $10,353 $1,000 $751 $12,103 10 $12,103 $1,000 $877 $13,981

44 See the above penalty table. Because the modification occurred during year six; the table estimates the total penalty at $7,199. To this we need to add an additional $1,000 of penalty for the extra $10,000 withdrawn in year six for a total of $8,199. The table assumes statutory interest is assessed at long-term applicable federal rate, which has averaged 7.25% over the past 10 years.

17

________________________________________________________________________________________________ Chapter 3 Other Strange Animals—Roth IRAs & NUA

Roth IRA Taxation and Penalties Thus far, our entire focus has been the elimination of the 10% surtax imposed by IRC §72(t) by searching for and implementing one or more of the available exceptions. In each of these cases, when properly implemented, the 10% surtax is avoided, but, more often than not, regular federal income tax (and potentially state income tax) is still imposed. Further, in this context, essentially all accounts (§401(a), §403(b), §401(k) plan balances and §408 IRAs) are treated as deferred accounts; the implica-tion being that taxation of the account proceeds is deferred to some point in the future, not eliminated or forgiven.

Roth Definition In the Taxpayer Relief Act of 1997, Senator William Roth championed a new type of IRA, accord-ingly called the Roth IRA45. For brevity purposes, we are going to shorten this to a “Roth.” The rules for Roths are completely different from all other types of IRAs. Roths have the following general char-acteristics:

• Unlike regular IRAs, no deduction is ever afforded for making a contribution. Further, the abil-ity to contribute to a Roth is generally limited to $2,000 per year46 for those taxpayers with less than $110,000 or $160,000 of adjusted gross income for single and married filing jointly tax-payers, respectively.

• Taxpayers with less than $100,00047 of adjusted gross income can also elect to convert some or all of their regular IRAs into Roths; however, this does create a taxable event on which regular income tax (but no penalty) is due48.

• Distributions from Roths are categorized as either qualified or unqualified. If the distribution is qualified, it is a tax-free distribution, irrespective of the composition or source of the distribu-tion dollars. If a distribution is unqualified, it is potentially taxable depending on a new set of special ordering rules discussed following.

Qualified and Unqualified Withdrawals Since there are many fine texts written all about Roth contributions, conversions, and recharacteri-

zations, we will skip past all of these issues and instead take a hard look at the distribution side of the

45 See IRC §408A and related regulations. 46 Commencing in 2002, IRA contribution limits, including Roth IRAs, will be increased $3,000; $4,000 in 2003; $5,000 in 2004 and indexed to inflation thereafter. 47 A Taxpayer may have AGI of less than $100,000 but may wish to convert IRAs of such size that their resultant AGI ex-ceeds the $100,000 limit. This is okay in that the AGI limit; imposed by IRC §408A(c)(3)(B) ignores the amount of the rollover or conversion in computing the taxpayer’s AGI. 48 On its face, a conversion from a regular IRA to a Roth is a withdrawal from the IRA and therefore subject to the 10% under §72(t). However, IRC §408A(d)(3)(A)(ii) gives us a break by saying that “§72(t) shall not apply”.

18

equation. Our objectives are, first and foremost, to completely avoid taxation and, second, if forced into a taxable event, to avoid the 10% surtax as imposed by §72(t). To do this we need to first under-stand the differences between qualified and unqualified distributions.

A qualified distribution is one that is BOTH:

• Made after a 5-taxable-year period, and

• Made on or after the date on which the taxpayer attains the age of 59½ , or made to a benefici-ary or the estate of the owner on or after the date of the owner’s death, or attributable to the owner’s being disabled, or for a first-time home purchase49.

An unqualified distribution is any distribution that is not qualified according to the rules above.

Application of 5-Taxable-Year Rule Based on the permissible types of transactions for a Roth, we potentially create three classes of as-

sets or dollars within the Roth account: annual contributions, conversions (from regular IRAs), and all earnings. Each of these asset classes receives different tax treatment upon distribution depending on the character (qualified or unqualified) of the distribution.

We already know a lot about exceptions 1, 2, 3, and 12; in fact the tests and rules are identical as those discussed previously, but what is a “5-taxable-year period”? Earlier, we discussed the concepts of “date specificity” and “highest attained age.” A 5-taxable-year period has a new set of rules. A 5-taxable-year period commences with the first contribution to a Roth account and is effectively back-dated to January 1st of the allocable tax year of the contribution and ends on December 31st of the fifth year. Actually, 5-taxable-year period could have been better phrased as “five complete taxable years, excluding the present tax year.” Examples include:

• January 1, 1998, was the first date ever that one could create a Roth and make a contribution. Let’s assume you created the account the account in March 1998 and made a $1,000 contribu-tion in June 1998 followed by another $1,000 contribution (allocated to the 1998 tax year) in February 199950. In this case, the “5-taxable-year period” commenced on January 1, 1998, and will conclude on December 31, 2002, five complete tax years later. Further, any subsequent year Roth contributions you make, say in 1999, 2000, and 2001, will all be afforded this same back dating for purposes of measuring the five years. Thus, the first date in history that you can make a qualified withdrawal is January 1, 2003, one day after the completion of the 5-taxable-year period.

• Assume the same facts as above, however, you didn’t get the word quickly enough and there-fore didn’t open a Roth account until March 2000. The 5-taxable-year period commences Janu-ary 1, 2000, and expires December 31, 2004. Thus, the first date to potentially make a qualified withdrawal becomes January 1, 2005.

Let’s now return to the tests for a qualified withdrawal. All one needs to do is to satisfy the 5-taxable-year rule AND be 59½ (or older) on the date of withdrawal, or be deceased, or disabled, or for

49 Interestingly, these rules seem to line up rather nicely to IRC §72(t)(2)(A)(i),(ii),(iii) and§72(t)(2)(F). These are the same as exceptions 1, 2, 3, and 12 as discussed in Chapter 2. 50 As a general rule, taxpayers have until April 15th (or later extended date of filing) of the year following to make an IRA contribution allocable to the prior year. This feature is afforded to Roths as well.

19

first-home purchase. When the withdrawal is qualified, the benefits are enormous: NO FEDERAL INCOME TAX. No tax now, no tax later, no tax never of any kind51. Thus, every taxpayer’s first objective should be to always attempt to make qualified withdrawals.

Contribution and Earnings Withdrawals What if you must take a withdrawal but fail the qualified test? All is not lost. Back in Chapter 2, we

discussed basis in account and the concept that withdrawals from regular IRAs could be part taxable and part tax free with the “basis” being withdrawn ratably with the taxable component. Well, here is a second gift thanks again to Senator Roth—unqualified withdrawals are ordered or sequenced such that 100% of a taxpayer’s contributions are deemed to have been withdrawn first, and only after all contri-bution dollars have been exhausted are earnings deemed to have been withdrawn. Again, some exam-ples:

• It is August 2002, and you are 45 years old. You created a Roth in 1998 and you have dili-gently made a $2,000 contribution for each tax year 1998 through 2002, totaling $10,000 in contributions. Through astute investing, the total account balance is now $22,000, of which $10,000 is treated as contribution dollars, and $12,000 is classified as earnings dollars. For whatever reason or emergency, you must make an $8,000 withdrawal. Although unqualified, this $8,000 withdrawal is a tax-free withdrawal because it is 100% comprised of contribution dollars on which you had already paid income tax. The $8,000 is not includible in gross in-come—therefore not taxed—and §72(t) does not apply.

• Assume the same fact set as above; the emergency, however, is of a nature that you must with-draw $12,000. In this case, the first $10,000 (all of your contribution dollars) are deemed to have been distributed first and are distributed tax free. The last $2,000 is another story. Because the withdrawal of the total amount was unqualified, the $2,000 becomes taxable (includible in your gross income). Not surprisingly, not only is the $2,000 taxed (at whatever is your then regular income tax bracket), but also §72(t) rears its ugly face again. Because the $2,000 is tax-able, §72(t) applies, and (unless one or more of the 12 exceptions apply) the 10% surtax, $200 in this instance, is enforced.

What can we learn from this? On the contribution side, Roths are a very powerful investment and asset accumulation tool. Done correctly they are tax free forever. This point is critical in comparison to all of the other deferred-asset accounts we discuss. In the later case, withdrawals are always going to be taxed; it is simply a matter of when and how much. Tax free is much better, and a Roth is currently the only place to get it. On the withdrawal side, first and foremost, always endeavor to make qualified distributions from a Roth!!! Second, if you must make an unqualified withdrawal, withdraw only amounts up to your lifetime-to-date contributions; these are also tax free. Only in the most catastrophic circumstance should one consider making an unqualified withdrawal of earnings dollars, because they will be taxed and—almost certainly—surtaxed, as one or more of the 12 exceptions is unlikely to ap-ply.

Further, assume for a moment that a taxpayer has both regular IRAs and Roths and has already withdrawn all of his or her lifetime-to-date contributions from the Roth. However, should an emer-

51 IRC §408A effectively says that a qualified withdrawal from a Roth is “not includible in the gross income of the tax-payer.” Therefore, the withdrawal is tax free, and IRC §72(t) does not apply by virtue that §72(t) applies only to “the por-tion of such amount that is includible in gross income.”

20

gency arise, from which account should the taxpayer withdraw? Without question, one should next withdraw from the regular IRA, not the Roth. Since the taxes and penalties are going to be the same for withdrawals from either account, the deciding factor is based on which dollars are more beneficial to protect. Clearly the Roth dollars are more precious in that one day they will become tax free, whereas the IRA dollars will always be taxed at some rate when withdrawn.

Conversion Dollar Withdrawals At the outset of the chapter we talked about three types of money in a Roth—contributions, con-

versions, and earnings. So far we have covered contributions and earnings. The rules for conversion dollars are again different. Whereas all contribution dollars are permitted to satisfy the 5-taxable-year period test commencing in the year the Roth was first opened and contributed to, this luxury is not af-forded to conversion dollars. Instead, each conversion receives it own discrete beginning date for satis-fying the 5-taxable-year rule. As an example, assume some similar facts as above: you made contribu-tions in 1998 through 2004; further, you also converted $10,000 each year from a regular IRA in 1998 through 2001. Finally, assume that the total account balance as of August 2004 is $90,000. Now a new set of ordering rules come into play:

• $14,000 of Roth contributions (1998 through 2004 inclusive).

• $10,000 of Roth conversions (1998).

• $10,000 of Roth conversions (1999).

• $10,000 of Roth conversions (2000).

• $10,000 of Roth conversions (2001)

• $36,000 of Roth inception-to-date earnings.

In a way, the Roth account starts to look like a layer cake with different taxability rules applying to each layer, again depending on whether a withdrawal is classified as qualified or unqualified. Remem-bering that you are 45 and it is now the summer of 2004, let’s see what happens under different cir-cumstances:

• If you make a withdrawal of $10,000 or less, it is a tax-free distribution, just as before.

• If you make a withdrawal of $30,000, you will be deemed to have withdrawn $14,000 of con-tributions, $10,000 of the 1998 conversion layer, and $6,000 of the 1999 conversion layer. We are still okay because in tax year 2004—both the 1998 and 1999 conversion layers have each independently satisfied the 5-year test. The 1998 $10,000 has been in the account for six full tax years (1998 to 2003) and the 1999 $10,000 has been in the account for five full tax years (1999 to 2003). Further, because the conversion dollars were taxed when you converted them from a regular IRA to the Roth, they can not be federally taxed again. Thus, all $30,000 comes out tax free.

• If you make a withdrawal of $54,000, you will be deemed to have withdrawn the $14,000 of contributions and all four $10,000 conversion layers. The good news is that the first $34,000 of withdrawals comes out completely tax free. Why? Because contribution dollars always come out first and tax free, and the 1998 and 1999 conversions have independently satisfied the 5-year test. What about the 2000 and 2001 conversions of $10,000 each? Because they are con-version dollars, they therefore already have been federally taxed and cannot be subject to re-taxation. However, and this is a big however, when each of the regular IRA layers was con-

21

verted to the Roth, IRC §408A(d)(3)(A)(ii) came into effect which says for “Rollovers from an IRA other than a Roth IRA…section 72(t) shall not apply.” This language needed to be present; otherwise the conversions from regular IRAs to Roths would be subject to IRC §72(t) and al-most certainly surtaxed. However, this is a temporary or conditional forgiveness of §72(t). IRC §408A(d)(3)(F)(i)52 is also applied. It says, in part, “if any portion of a distribution from a Roth is properly allocable to a qualified rollover contribution [a “conversion” in our vocabulary]… and such distribution is made within the 5-taxable-year period…then §72(t) shall be applied as if such portion were includible in gross income.”

What does this mean? Since each conversion had its own independent time clock, in 2004 the 1998 and 1999 conversions had surpassed the five-year test; the 2000 and 2001 conversions, however, are too new and have been in the account less than five full taxable years. As a result, §408A(d)(3)(F)(i) says that we need to reapply §72(t) as if those distributions were taxable in-come. We know that the $10,000 conversions are, unto themselves, not taxable because they have already been taxed when the money went in; this code section, however, is saying that we need to momentarily pretend that they were taxable for §72(t) purposes. As a result, it is ex-tremely doubtful that any of the other §72(t) exceptions will apply; therefore, a §72(t) surtax of $2,000 will be assessed.

• Let’s take the last and worst case: you withdraw all $90,000. The first $34,000 comes out tax free; the second $20,000 comes out tax free, but is surtaxed $2,000; the last $36,000 comes out and is taxable (let’s assume at 28%) and is further surtaxed at 10% for an additional $3,600. Your total tax bill is $15,680, of which $5,600 are §72(t) penalties.

In summary, Roths are a very powerful asset accumulation tool, particularly for the younger in age as well as those with sufficiently modest adjusted gross income that you are able to make both contri-butions and conversions. However, the withdrawal rules, as demonstrated above, are different than all other IRAs and are not straightforward. Any taxpayer under age 59 ½ and who has made regular to Roth IRA conversions should be wary of making withdrawals; it is just too easy to forget that each conversion has its own time clock of five full tax years and a withdrawal of a conversion layer before the five year mark will have the 10% penalty reimposed.

Net Unrealized Appreciation (NUA) If the third word in NUA were “appreciosis,” everyone would be convinced it was a terminal dis-

ease. Fortunately, it’s not. Instead, it is very powerful tax avoidance54 tool but unfortunately limited to select taxpayers.

52 On the one hand, these “pretend to look back” rules appear to be convoluted and downright silly. On the other hand, they are logical from a different perspective. IRC §72(t), which was on the books for several decades before §408A, makes it difficult to get money out of a deferred account before age 59½. Assume you are in your 40s or early 50s and wished to make withdrawals from your regular IRAs but did not qualify for any of the exceptions. All else being equal, you could certainly do so but would be subjected to the §72(t) penalty of 10%. However, were IRC §408A(d)(3)(F)(i) to be absent, one could easily convert periodically as needed from a regular IRA to a Roth at will and then make withdrawals from the Roth with no penalty, essentially circumventing §72(t). 53 There is no footnote 53. However, anyone who reads this fine print and sends an email to themarblegroup@att.net will automatically be entered into the database to receive a free copy of this report, third edition, due out in the fall of 2002. 54 Please remember that “tax avoidance” is always very legal and sometimes fun; “tax evasion” is always illegal. It too may be fun, at least until caught.

22

Many taxpayers belong to qualified plans sponsored by an employer. Further, many employers make profit sharing or matching contributions to deferred plans in the name of a employee. Most often these employer contributions are made in cash; sometimes the contribution is made “in kind.” If it is an in kind contribution, it will invariably be shares of employer common stock55 or other related employer security. Further, even if the contribution is made in cash, the employee may be afforded the opportu-nity to invest those cash proceeds directly in company stock. Some employers further permit an em-ployee to allocate voluntary contributions to the purchase of company stock56.

Over a period of years, it is quite conceivable that an employee might acquire a material number of shares in the company. Further, longevity with the same company gets rewarded in that the employee may acquire thousands of shares at relatively low purchase prices.

Let’s try an example to see if this concept becomes clearer. Assume Bob, age 52, has been working for First United National (symbol: FUN) for 30 years. During his 30-year tenure, he has acquired 7,000 shares of FUN currently valued at $80 per share or $560,000 total. In addition, Bob also has another $140,000 in his deferred account invested in mutual funds. Those 7,000 shares were each purchased in very little layers over the 30 years—such that the first share he bought 30 years ago cost him $1 per share, and the last two shares he bought (last week from his payroll deduction to his §401(k) plan) cost him $80 per share. Keeping track of 500 or more individual little purchases is rather onerous for a plan administrator particularly when there are 500 or 50,000 employees to track. The plan administrator does not track each purchase; instead, a cumulative purchase price bucket is maintained for each em-ployee which represents the total dollars expended to purchase all shares over the 30 years. In our case, let’s further assume that Bob has paid $140,00057 for his cumulative 7,000 shares, or, on average, $20 per share. Another way to express this would be to say that Bob has a purchase basis of $20 per share.

FUN is trading today at $80 per share. Bob’s NET UNREALIZED APPRECIATION is $420,000 (($80 - $20) * 7,000).

This is good news. The bad news is that a month ago, Bob’s boss informed him the FUN no longer needed his services, and Bob is separating his employment with FUN today. Further, the provisions of the FUN retirement plan either permit or require that all proceeds in the plan be immediately distrib-uted upon termination of employment. What should Bob do?

Actually, Bob has two fundamental choices. First, Bob can ask that a check be prepared for a total of $700,000 and sent directly to his brokerage of choice and deposited into the “Rollover IRA Account for the Benefit of Bob.” This transfer, actually a trustee-to-trustee transfer, is a nontaxable event, and the $700,000 now in the rollover IRA takes on all the characteristics of any traditional IRA. Thus, Bob will be fully taxed on any withdrawals at regular income tax rates (let’s assume a combined federal and state rate of 30%). Bob, being the smart fellow that he is, is fully conversant in §72(t) issues and estab-lishes a substantially equal periodic payment stream that pays him $60,000 per year and avoids all §72(t) surtaxes. Thus, one way to look at this option is to say that his IRA has a net after-tax value of $490,000 (70% * $700,000). 55 Needless-to-say, this can be a good event or a bad event depending on whether the stock price goes up or down, respec-tively. 56 This author is neither an advocate nor an opponent of employees holding material amounts of employer stock in deferred accounts; rather, these situations simply exist. How the taxpayer got there is not particularly relevant. 57 It matters not whether Bob’s contributions or FUN’s contributions to Bob’s account were used to purchase of the shares.

23

Bob’s second option is to avail himself of the special (and favorable) tax treatment afforded to the distribution of employer securities from a qualified plan58, generally called the NUA rules. These rules permit Bob to take a qualifying lump-sum distribution from the FUN plan, directing the 7,000 shares be distributed to him directly and the remaining $140,000 of in mutual funds be sent to his rollover IRA59. In this case, a tax-free transfer of the $140,000 occurs, and a taxable distribution of the 7,000 shares to Bob occurs. Since Bob is not yet 59½, it looks like he is going to have to pay regular income tax on the 7,000 shares plus the 10% surtax (how disgusting). However, the silver lining is that Bob receives the 7,000 shares at his purchase basis of $140,000 (7,000 shares * $20 per share) not the cur-rent market value of $80 per share. So, Bob has a taxable distribution of $140,000, upon which he pays $42,000 in taxes ($140,000 * 30%) and $14,000 for the §72(t) surtax of 10% (because none of the §72(t) exceptions apply). Next, Bob deposits the 7,000 FUN shares in the brokerage of his choice and sells them and is afforded long-term capital gains tax treatment. His basis in each share was $20; he sold for $80 a share. His long-term capital gain is $420,000, on which he pays another $84,000 in taxes at 20%. For comparative purposes, assume that that other $140,000 in mutual funds has an after-tax value of $98,000 ($140,000 * 70%).

Which plan is better plan for Bob? Under option #1, Bob had an effective after-tax value of $490,000. Under plan #2, Bob has an after-tax value of $518,000 ($700,000 received minus $42,000 of tax on the $140,000, minus $42,000 of tax on the 7,000 shares, minus the $14,000 of §72(t) surtax, mi-nus the $84,000 of long-term capital gains tax). Option #2, irrespective of the §72(t) surtax, results in $28,000 more in Bob’s pocket—thus the winning strategy, primarily because of the tax rate differential of 10% (in this example) between the tax rate on ordinary income versus the tax rate on long-term capital gains.

What if Bob’s basis in the 7,000 shares were $60 a share instead of $20 per share? Option #2 would turn into a losing strategy with only a $462,000 after-tax value. The critical issue here is the re-lationship between the size of the unrealized gain in the shares compared to Bob’s regular tax bracket. We can develop a break-even table as follows:

NUA Break-Even Mutliplier Table Your Federal Tax Bracket Multiplier

15% 3.00 times 27.5% 2.33 times 30.5% 1.95 times 35.5% 1.65 times 39.1% 1.52 times

In summary, if you are currently in the 27.5% federal tax bracket, the current market price of your FUN shares must be 2.33 times as your average basis in those shares for the NUA treatment to work beneficially; e.g. in our example we can take the current trading price of FUN @ $80 per share and di-vide by 2.33 to get $34.33. Therefore, as long as your basis is under $34.33, it will be beneficial to use

58 See IRC §402(e)(4) Net Unrealized Appreciation and Related Regulations. 59 One of the NUA rules is that all the assets of an employee’s account must be distributed in a qualifying lump-sum distri-bution; the destinations and tax status of those assets may, however, vary.

24

the NUA treatment. Similarly, if you are in the 35.5% tax bracket60, your basis in the FUN shares need only be below $48.48 for NUA treatment to work. As usual, there are a variety of rules and traps:

• NUA treatment can be used only when the distribution from the plan, in which you are a mem-ber, is a total qualifying lump-sum distribution. Generally, this means all of the assets in the plan that are yours are distributed, and no assets are left in the plan to which you have any claim of ownership.

• NUA treatment is afforded only to employer securities61. This means that because you worked for FUN, it only applies to FUN shares and does not apply to any other securities you might own within the plan. Further, these shares must be real shares of stock or other types of real employer securities that are specifically allocated and tracked within your plan account; trust shares, mutual fund shares, and shares of prior employers are all ineligible.

• As with many taxpayer elections, NUA treatment is afforded only prospectively and must cre-ate a taxable event. As an example, if you roll the shares over to an IRA, the game is over. You cannot back up and take the shares into a taxable account after the rollover and receive NUA treatment.

• The multiplier table above is accurate for 2001, but only under certain conditions: one, the NUA taxable event does not cause a jump upwards in your marginal tax bracket; two, zero state taxes are assumed; and three, a capital gains rate of 20% is assumed for all brackets—except for the 15% bracket, where a 10% capital gains rate is assumed62.

Three or four pages are totally inadequate to provide a comprehensive treatment to the NUA sub-ject. Above are just the basic rules with just enough specificity in the examples so the reader will get a reasonable feeling for how the process works. The point of this section is to help you evaluate your quarterly plan statements to see if your shares in company stock warrant NUA treatment. When ready, you should visit with a trained tax accountant or tax attorney. As usual, there is one right way and a dozen wrong ways to make this election. Further—as with much of the subject material in this text—when an error is made, there are few, if any, ways to back up and correct the mistake.

Summary

In this chapter, we covered two situations: Roths including their tremendous advantages as well as a few disadvantages and the basics of the NUA rules. Further, we can quickly revert to Chapter 2 and pull in exception #5, Separation of Service at Age 55. All three of these circumstances have a common thread—the need to look and plan ahead, sometimes a couple of months, sometimes five years to

60 Through tax year 2001, the above table of multipliers was accurate. Commencing 2002 through at least 2006, the mar-ginal tax brackets will be dropping gradually, typically a half or whole percent annually. As a result, there is no way to keep this multiplier table accurate. Individual taxpayers in these circumstances are strongly urged to model out the taxes paid component for the then current tax year to obtain precise and accurate results. 61 Employer securities can be common stock, preferred stock, bonds, debentures, etc. However, it is typically only the common shares, convertible preferred shares, and convertible debentures that realize sufficient appreciation after purchase to overcome the 1.5 to 3.0 multiplier to make the NUA treatment financially positive. 62 When either a jump in your federal tax bracket will occur or state taxes will apply, it is best to ignore the multiplier table, other than as a guide post and, instead, calculate the actual numbers involved including development of rough proforma tax returns.

25

decades in advance. None of these exceptional situations is really difficult to implement although some require a touch of professional help. More importantly, they are easy to miss and then become irrevo-cably lost the instant a taxpayer takes these assets and rolls them over to an IRA.

This author’s recommendation is that anyone contemplating retirement, actually at any age, should start the investigatory process at least one year in advance of the anticipated exit date. This then allows for a full examination of all alternatives at a leisurely pace.

26

_______________________________________________________________________________________________ Chapter 4 Computation of Substantially Equal Periodic Payments It would be convenient if a taxpayer could select any amount he or she desired and simply keep the withdrawal amount equal from year to year. Unfortunately, that is not the case. IRS Publication 590 (see Appendix D for an excerpt) provides some guidance; however, it is incomplete63. To quote in part, Publication 590 informs the taxpayer that:

But for the life expectancy method (sometimes called the minimum method), Publication 590 di-rects the taxpayer to Notice 89-25, 1989-1 Cumulative Bulletin 66264 (see Appendix C for an excerpt). Within Q-12 of the notice, we finally get to the heart of the matter and a reasonably detailed explana-tion of the three currently acceptable computational methods. They are the “minimum,” the “amortiza-tion,” and the “annuity” methods.

Because computations are easiest to explain via example, we are going to build an assumed fact set for John Q. Taxpayer and use these facts consistently throughout the following examples.

• John is age 52 during 1999 and is married to Cathy, age 50 in 1999, who is his beneficiary.

• As of December 31, 1998, John has a total of $1,000,000 in four rollover IRAs: A with $400,000; B with $300,000; C with $200,000, and D with $100,000.

• Although Cathy has deferred-asset accounts as well, she has elected to not make any early withdrawals.

• John terminated his employment with XYZ Corp. effective December 31, 1998, and wishes to commence SEPPs in 1999 as an income replacement.

• For simplicity, neither John nor Cathy are disabled, and neither will become deceased in the next 10 years.

63 Actually, for Publication 590 to use the language “must use” is actually incorrect, or at a minimum represents the IRS taking typical broad poetic license with its publications. Notice 89-25 gives us an affirmative statement: “payments will be considered to be..[SEPPs]...if they are made according to one of the methods....” Nowhere in the IRC or Notice 89-25 is there any language to the effect that there are only three acceptable methods to the exclusion of others. Thus, it is entirely possible that there are methods four, five, and six, which are equally acceptable. However, none of these other methods have been developed or proposed in any PLRs or other documents of authority. 64 As an editorial aside, this writer objects to the IRS’s whole handling of §72(t) issues. Most, and maybe virtually all, tax-payers have little idea what a Notice is much less any understanding of what a Cumulative Bulletin is or where to find one. One could almost come to the conclusion that someone has made figuring out SEPPs intentionally difficult. As a result, we have attempted to make this text as inclusive as possible such that the average taxpayer will have everything at his or her immediate disposal in the Appendices.

“You must use an IRS-approved distribution method and you must take at least one distribution annually...for one IRS-approved distribution method, generally referred to as the ‘life expectancy method.’...There are two other IRS-approved distribution methods that you can use. They are generally referred to as the ‘amortization method’ and the ‘annuity factor method.’”

IRS Publication 590

27

Minimum Method This method is described in Notice 89-25 as follows:

The critical language here is “acceptable...under §401(a)(9).” IRC §401(a)(9) and related regula-tions are primarily oriented toward the computation of minimum distributions for individuals who have attained the age of 70½. However, Notice 89-25 is telling us to essentially pretend and apply all of the same rules and methodology contained in the regulations as if John has attained the age of 70½.

Substantial amounts of discussion exist on the minimum method within Publication 59065 as it is reasonably easy to understand. The taxpayer will need life expectancy tables, as published by the IRS. They can be found in Appendices E and F of this text as well as in Publication 590, Appendix E, Life Expectancy Tables. Table I is for a single life; Table II is for the joint lives of the recipient and the re-cipient’s beneficiary66. In our case, John at age 52 has a life expectancy of 31.3 years (from Table I). John and Cathy, aged 52 and 50 respectively, have a joint life expectancy of 38.3. John has a statutory right to chose either the single or joint life expectancy value based on the statute’s use of the term “made for the life (or life expectancy) of the employee or the joint lives (or joint life expectancies) of such employee and his designated beneficiary.”

Thus, if John chooses the single life expectancy, his SEPP for 1999 will be $31,949 ($1,000,000 / 31.3). If he elects the joint life expectancy, his SEPP for 1999 will be $26,110 ($1,000,000 / 38.3). These results are based on the assumption that John also chooses to include all four of his IRAs in his SEPP universe.

Moving forward, in the second and subsequent years John has a choice to make. He can (assuming he chose the single life expectancy value) keep his year two and subsequent distributions fixed67 at $31,949 per year, or he can choose to recalculate. To do so, we need to determine John’s deferred-asset balances as of 12/31/99. Let’s assume they grew by 10% to $1,100,000 less either the $31,949 or $26,110, whichever John elected. Further, we need updated life expectancies68 for John alone (30.4 for a 53-year-old) and John and Cathy (37.3 for a 53- and 51-year-old). As a result, John’s SEPP in 2000 would be either:

65 Publication 590 is considerably easier to read than IRC Reg. §1.401(a)(9). Of particular relevance are pages 21-25 in the 1999 version; pages 18-21 of the 2000 version. 66 These same tables are published as Tables V and VI in IRC Reg. §1.72-9. Unfortunately, IRC Reg. §1.401(a)(9)-1: E-3 and E-4 @ ¶ 17,724 unequivocally force us to use these tables to the exclusion of all others. 67 In this case, John does have the option (a one-time option at the commencement of SEPPs) to either fix the amount or annually recalculate. This would not be true were John age 70½, as then the minimum distribution rules would essentially force annual recalculation. 68 We are assuming that John (and Cathy as needed) elected to recalculate their respective life expectancies annually.

“Payments shall be treated as satisfying section 72(t)(2)(A)(iv) if the annual payment is determined using a method that would be acceptable for purposes of calculating the minimum distribution required under section 401(a)(9). For this purpose, the payment may be determined based on the life expectancy of the employee or the joint life and last survivor expectancy of the employee and beneficiary.”

IRS Notice 89-25

28

1. $35,133 ( ($1,000,000 - $31,949 + $100,000) / 30.4) using a single life expectancy; or 2. $28,791 ( ($1,000,000 - $26,110 + $100,000) / 37.3) using joint life expectancies69.

This process of annual recomputation70 would continue until John (ignoring Cathy’s age even when using the joint life expectancies) and only John attains the age of 59½ .

Advantages—First, on the presumption that John is a reasonably astute investor, is that John will receive a “pay increase” each year by virtue of the numerator increasing (by investment gains in excess of the amounts withdrawn) and the denominator decreasing (by virtue of the decreasing life expec-tancy(ies)). Second, on the downside, but nonetheless an advantage, no matter how poorly John might invest, he can never exhaust the assets of his deferred accounts, although the latter year minimum dis-tributions might become very small.

Disadvantages—First is that this method delivers a relatively low annual distribution amount (compared to other methods discussed following). For early retirees, 1/30.4 is approximately 3.3%. Second, the minimum method permits, but does not require, annual recomputation. This requires the careful scrutiny of documents from multiple trustees (in the case of multiple accounts) and recomputa-tion of life expectancies. Although not particularly difficult in a mathematical sense, the potential for error exists, which could result in an incorrect mathematical result71.

As a general comment, there are very few PLRs relating to the minimum method. Those PLRs that do exist in conjunction with the minimum method were invariably issued due to a rather unique tax-payer fact set that was targeted toward issues other than the method selected. Further, there has been little need for PLRs on the minimum method because we essentially have a surrogate regulation in §1.401(a)(9) as well as pretty clear instructions in Publication 590.

69 To the extent that John elects to use joint life expectancies, John should make sure that Cathy, in this particular example, is in fact the named beneficiary on the account(s) in question. See §1.401(a)(9)-1: D-1 and D-2. 70 As we will see later, John may recalculate annually, but must do so using the same table, i.e. single life expectancy or joint life expectancies. He cannot switch tables between years. Nonetheless, here is a specific circumstance where the IRS, by its adoption of the method, has deemed mathematically differing amounts by year to be “substantially equal.” 71 This author has found no ruling or other circumstance where the taxpayer was dutifully following this method, or either of the two others, and made a mathematical error resulting in a subsequent year incorrect amount. On its face, such an amount would fall outside the umbrella of deemed substantially equal. What the IRS would do in such a circumstance is still unknown. Still, there is no reason to temp fate. Every taxpayer should re-check his or her math twice, if not three times.

29

Amortization Method Notice 89-25 describes the amortization method as follows:

With the amortization method, we are faced with making at least two decisions: one, choose be-tween single and joint life expectancies72, and two, select an interest rate. Again, we’ll use John and Cathy Taxpayer in our example. For life expectancies, we use the same ones as described in the mini-mum method, 31.3 (single life) and 38.3 (joint lives) years respectively. Later, we will discuss the se-lection of interest rates, but for this example we will assume an interest rate of 8%. What annual pay-ment amount will fully amortize73 (meaning reduce to zero) the original $1,000,000 over 31.3 or 38.3 years assuming an interest rate of 8%? Either $87,904 or $84,430 depending on whether John chooses the single life (31.3) or joint life expectancies (38.3). In the author’s opinion, the easiest way to derive these values is through use of the “@PMT” function in either Lotus or Excel. The syntax of the function is: @PMT(P,I,T); where “P” is the beginning principal amount, “I” is the annual interest rate, and “T” is the term expressed in years. As an example, if one wishes to amortize $100,000 over 20 years at 8%; it would be expressed as @PMT(100000,.08,20) with a resultant answer of $10,185.22. For those individuals with no access to an electronic spreadsheet product, the same can be accomplished using any “financial series” calcula-tor made by Texas Instruments, Hewlett-Packard, as well as others. Please note that the minimum method required annual recalculation. Recalculation when using the amortization method is an open question dealt with in great detail later in the text. However, for the moment, we will assume that no recalculation is performed, e.g. a very vanilla interpretation of Notice 89-25. Advantages—The amortization method has its own set of advantages. First, the computation can yield annual distribution amounts substantially higher than the minimum method. Whereas the mini- 72 As indicated in Notice 89-25, the amortization method mandates the use of Tables V or VI for life expectancy determina-tion. If we leap forward for just a moment to the annuity method, we will find considerably different language on this issue. The annuity method allows “using a reasonable mortality table,” which is considerably more flexible than the amortization method. Is the language inconsistency between the two methods accidental or intentional? Unfortunately, no one knows. All PLRs involving the amortization method used Tables V or VI; no one has requested a PLR using the amortization method and any tables other than V or VI. 73 One can think about the amortization method as the same as a home mortgage, e.g. I need to borrow $1,000,000 from the bank at 8% for 31.3 years. What is the annual payment amount?

“Payments will also be treated as substantially equal periodic payments within the meaning of section 72(t)(2)(A)(iv) if the amount to be distributed annually is determined by amortizing the taxpayer's account bal-ance over a number of years equal to the life expectancy of the account owner or the joint life and last survivor expectancy of the account owner and beneficiary (with life expectancies determined in accordance with pro-posed section 1.401(a)(9)-1 of the regulations*) at an interest rate that does not exceed a reasonable interest rate on the date payments commence.”

IRS Notice 89-25

* Reg. 1.401(a)(9)-1 is a rather lengthy regulation; the essence of which, for our purposes, forces us to use Table V or Ta-ble VI for life expectancy purposes in conjunction with the amortization method. Further, there are approximately 40 PLRs using the amortization method, each dealing with unique fact sets as well as other questions of interpretation; however, in all 40 PLRs each taxpayer elected to use either Table V or Table VI. No one to date has suggested or requested the use of any other life expectancy tables.

30

mum method provided annual distributions in the 3.3% range, the amortization method provides an annual distribution in the 8.5% range given John’s age and the presumption of an 8% interest rate. Second, this computation is usually done once and only once, thus eliminating most possibilities of mathematical error. Third, where the minimum method is fixed (other than life expectancy table selec-tion), the amortization method offers us some additional flexibility in the ability to select differing in-terest rates. As an example, if John chooses 7% instead of 8%, the single life withdrawal amount drops to $79,573, approximately 10% less per year. Disadvantages—On the other hand, the amortization method does not offer the built-in “pay in-crease” as found in the minimum method. Second, there is a risk that John, through unwise invest-ments, could potentially deplete the entire principal amount before attaining the age of 59½, thus pre-maturely exhausting his IRAs. There is more discussion on this “account exhaustion” exposure later in this document in Chapter 5.

Annuity Method The annuity method is described in Notice 89-25 as follows:

Note the language difference here between the annuity method and the amortization method. Whereas the amortization method forced us to use either Table V or VI, the annuity method provides considerably more latitude in the selection of a “reasonable mortality table.” There is considerably more on this issue later in the text. The mathematics involved here is a bit more sophisticated74. To implement the annuity method, it is best to engage the services of an actuary or an accountant familiar with annuitant functions and annuity calculators. Or, if the reader is so inclined, we have described how to build your own annuitant factor calculator in Chapter 6, devoted to life expectancy table selection. In our case, the annuity factor for John (age 52) assuming 8% is 10.845275. Thus his annual distri-bution would be $92,207 ($1,000,000 / 10.8452). Advantages—The advantages and disadvantages of the annuity method are very similar to those of the amortization method, with one addition. The annuity method will always yield a higher SEPP amount commencing at approximately age 44 with a widening spread approaching 10% more than the

74 It is the SUM from the recipient’s current age to age 115 of the probability of attaining current age plus 1 times the value of $1 discounted by 1 plus “n” periods. Don’t worry about untangling this sentence, we will show you step-by-step how to build an annuitant divisor table a little later in Chapter 6. 75 See Appendix H, last page, last row. Additionally, this all seems like black magic at the moment; e.g. where did 10.8452 come from. Unfortunately, there is no quick math here; instead we need to go through the details of really building an an-nuitant divisor table which is covered in great detail in Chapter 6.

“Finally, payments will be treated as substantially equal periodic payments if the amount to be distributed annu-ally is determined by dividing the taxpayer's account balance by an annuity factor (the present value of an annu-ity of $1 per year beginning at the taxpayer's age attained in the first distribution year and continuing for the life of the taxpayer) with such annuity factor derived using a reasonable mortality table and using an interest rate that does not exceed a reasonable interest rate on the date payments commence.”

IRS Notice 89-25

31

amortization method at age 58. Thus, were John’s objective to maximize his annual SEPP amount, he would chose the annuity method.

Disadvantages—The additional disadvantage is that the annuity method is the most difficult of the three to compute and may require the assistance of a professional.

Other Methods As of this writing, there are no other approved methods. However, that does not necessarily mean that there might not be some in the future. As an example, John might very specifically want to com-pletely exhaust his total IRA by age 7076. Currently, at age 52 and assuming 8%, his annual distribu-tion would be in the range of $104,000 to $107,000 depending on method selected—a considerably larger number than before.

76 John’s family has a long history of short-lived males, and John wishes to exhaust his IRAs quickly, leaving only regular assets (which receive a step-up in basis) to his heirs.

32

_______________________________________________________________________________________________ Chapter 5 Interest Rate Selection Risks External Risk Notice 89-25 contains some specific language regarding interest rates: “…at an interest rate that does not exceed a reasonable interest rate on the date payments commence77.” This language is present in both the amortization and annuity method explanations78. Thus, we are given two pieces of guid-ance: one, the IRS is apparently concerned about taxpayers selecting an interest rate that is too high through it’s use of the “does not exceed” language; two, the date of measurement to determine the rea-sonableness of an interest rate is the date payments begin (not necessarily the same as the date of calculation or account valuation). Almost all of the affirmative private letter rulings on §72(t)(2)(A)(iv) issues contain some very similar language:

“The life expectancies and the interest rate used are such that they do not result in the circumvention of the requirements of §72(t)(2)(A)(iv) and §72(t)(4) of the Code (through the use of an unreasonable high interest rate or an unreasonable life expectancy)79.”

Based on the above, we can reasonably presume that the IRS is not too concerned about a taxpayer selecting an interest rate that is too low. The reason for this is that the taxpayer has many tools at his or her disposal to solve a too-low rate, e.g. splitting a big IRA into two smaller ones or using joint and several lives instead of single-life computations. Unfortunately, Notice 89-25 falls short of giving us a nice clear-cut definition, like the in-force rate of a 10-year Treasury Note plus 150 basis points. In-stead, we are left to our own devices, a guessing game as to what constitutes an interest rate that is rea-sonable80.

As an aside, this author believes that each taxpayer should always assume that he or she will be audited and that such audit will always uncover the weak points in a taxpayer’s return. Thus, what is the downside financial risk if the taxpayer selects an interest rate that exceeds a reasonable rate in the eyes of the IRS? This downside risk is either the costs associated with the imposition of the 10% surtax and intervening statutory interest from the commencement of the payments, or the costs incurred to prevail over the IRS’s challenge in a legal proceeding, or the costs associated with losing the litigation plus having the surtax and interest imposed.

Thus, we are faced with a significant external risk of selecting an interest rate that is too high, but not too low. Fortunately this risk is manageable, as we will discuss in the following sections.

77 This is about as clear as mud. 78 Notice 89-25, page 8, A-12. 79 As a sampling for the interested reader, see private letter rulings 98-01050, 98-12038, 98-16028, 98-24047, 98-30042, and 99-09059. 80 The IRS has repeatedly refused to provide written and public guidance on reasonable interest rates other than by implica-tion imbedded within a PLR. On the other hand, since this issue of reasonable interest rate has never been litigated, it would presumably rest on the shoulders of the IRS to demonstrate that an interest rate selected by a taxpayer was unreasonable in order to void the protection of §72(t)(2)(A)(iv).

33

Internal Risk We also face an internal risk called account or asset “exhaustion.” This risk diminishes with age such that it is the greatest in the early years of substantially equal periodic payments and diminishes as one approaches age 59½. It is conceivable that a taxpayer could commence SEPPs at any age before 59½ with any beginning account balance and any set of factors and interest rate selected in conjunction with the amortization or annuity method. As a result of these decisions, coupled with poor market and investment returns, it is entirely conceivable that this taxpayer’s account could be exhausted or de-pleted of all assets before satisfying the greater of five years or the age 59½ rule. As a result of this as-set depletion, the taxpayer would become literally and practically unable to satisfy the remainder of the required SEPP stream of withdrawals. If this happens, what might the IRS say? We would all hope that the IRS would show some mercy on this taxpayer. After all, he has just lost all the assets in the SEPP account; what more can the government ask? Unfortunately, this issue has never been raised in any publication of the IRS, official or unofficial, and has never been litigated. As a result we have no guid-ance other than common sense, which is not necessarily the same as mercy. IRC §72(t)(4) is the code section that reimposes the 10% surtax and interest back to whenever the SEPPs began. Further, the rules causing the implementation of these penalties are very straightforward. Unfortunately, asset exhaustion is not one of the exceptions. As a result, although it appears harsh, this author is convinced that if a taxpayer suffers complete asset depletion, the IRS will impose the penal-ties of §72(t)(4) for several reasons:

1. Neither Congress nor the IRS is responsible for what would be called the imprudent acts of a tax-payer and, therefore, are not obligated in any fashion to grant that taxpayer any relief from the law.

2. The IRS would most likely be prevented from granting some mercy even if it were so inclined since the imposition of the 10% surtax and interest is statutory in nature81. As a result, the IRS has no room to maneuver. In this case, Congress has spoken directly through legislation and enactment of §72(t)(4). The IRS has no authority to circumvent the law, which is how a grant of leniency could be interpreted82.

3. Granting relief to a taxpayer suffering from asset exhaustion could also represent a moral hazard, e.g. the mathematically facile could always construct a unique set of financial circumstances where it could literally become beneficial for a taxpayer to intentionally exhaust a SEPP account thus ceasing SEPP payments.

As a result, in addition to the external risk discussed previously, we also face an internal risk of ac-count exhaustion. This risk is also manageable, as we will discuss below. 81 In this particular case, the imposition of the 10% surtax was legislated and is therefore contained in a specific IRC code section, unlike other issues, which Congress will frequently leave to the discretion of the IRS with respect to detailed deci-sion making, interpretation, and implementation. 82 In an unrelated General Counsel Memorandum, (think of this as a legal brief from your lawyer), the IRS asked its own lawyers to study the issue of the 60-day rollover rule as it pertained to IRAs. In this particular fact set, a number of taxpay-ers were defrauded of IRA assets by a crook posing as a qualified trustee with a bogus investment scheme resulting in all investors losing all of their invested assets. The IRS appropriately said that the investors had improperly transferred IRA monies (because the crook was not a qualified trustee) but that each investor could avail himself to the 60-day rollover rule to replenish the original IRA assets. Unfortunately, since more than a year had passed, the IRS asked its general counsel for some leniency in permitting these taxpayers more than the 60 days. The answer was no because the 60-day rule is statutory (meaning drafted by Congress), and the IRS had no authority to override congressional language irrespective of the agre-giousness of the circumstance. Here is a circumstance where the IRS was attempting to help taxpayers and failed due to language in the IRC. This author strongly suspects that the outcome would be same for a taxpayer who has inadvertently exhausted his IRA assets prematurely.

34

Internal Risk Management There are a variety of factors that can influence the taxpayer when selecting an interest rate to be used in conjunction with either the amortization or annuity methods. On the positive side, an increased interest rate will increase the annual distribution per dollar of underlying assets. For example: A 50-year-old using the amortization method at 7% will receive an annual distribu-

tion of $7,834 per $100,000 of assets. Increasing the interest rate to 8% causes a corresponding rise in the annual distribution to $8,679, an increase of approximately 10.8%.

As the interest rate selected increases, the relative change or marginal increase in the distribution amount decreases such that interest rates above 10% create distribution increases in only the 7% to 8% range per 1% absolute change in interest rate. The effect of interest rates on the annuity method are similar but somewhat smaller, averaging an 8% marginal change in distributions with interest rates un-der 10% and correspondingly lower marginal returns when the interest rate is above 10%. All this be-ing said, for some a 8% to 10% annual increase in distributions will be significant; for others it will be immaterial. On the negative side, lower commencement ages and higher interest rates increase the risk of ac-count exhaustion before age 59½. With all else equal, it would appear to be much safer for a 55-year-old to select an interest rate of 10% than for a 45- to 50-year-old to do so.

For example: Sally is 55 and selecting the annuity method at 10%. Her annual distribution will be $11,024 per $100,000 of assets. Very simply, Sally need withdraw only approximately 55% of the total underlying assets to reach the safety zone at age 59½. Thus, her probability of exhausting the underlying assets is not impossible, but certainly very low on a comparative basis. Conversely, Edward is a 45-year-old selecting 10%. His distribution will be $10,044 per $100,000 of assets. Edward is looking toward a 15-year program of distributions totaling almost $151,000, which is 151% of his underlying assets. Said another way, Sally can withstand a negative 17% return per year on the underlying assets in the account and still have some pocket change left at age 59½ . Conversely, Edward must achieve a minimum 6% positive return per year. To not do so will result in the exhaustion of the underlying assets before he reaches age 59½. Further, any negative invest-ment returns in the first several years of the SEPP program can do nothing but acerbate the prob-lem.

Just how sensitive are SEPPs to the interest rate selected? The relationship between risk and interest rate is a linear one. For example: Let’s return to example of Edward, our 45-year-old above. If he selects 6% instead

of 10%, the annual distribution drops to $7,089 per year per $100,000 of underlying assets. Fifteen years of programmed distributions equals approximately $106,000 or 106% of underlying assets, down from 151% but still a relatively high number. A 40% decrease in the interest rate caused a corresponding risk decrease of just under 30%. Instead, if we decrease the program period by 40% to 9 years (down from 15 years) and hold the interest rate at 10%, the annual distribution increases to $10,556 per year or $95,000 of total program distributions or 95% of the underlying assets. Ac-cordingly we can see that the risk factor decreased by 37%. A linear relationship is apparent with a slightly higher correlation than the interest rate relationship.

We have established that the risk of account exhaustion has a dual linear relationship with age and interest rate. To be useful, however, a risk measurement needs to be relative. At the end of this chapter is a table of account exhaustion percentages based on age and interest rate using the amortization method.

35

For example: Looking at the table, a 47-year-old using 7% will need to plan on program distribu-tions equal to 99.79% of beginning underlying assets.

At what point should we consider the relationship of age and interest rate to be too risky? To this question there is no single correct answer. As a simple suggestion, any intersection above 100% might be considered as too risky, and any intersection below 100% is acceptable. This is not necessarily the case for each taxpayer. The risk-prone taxpayer might decide that accepting 110% or 120% is okay; conversely, the risk-adverse taxpayer might decide not exceed 80%. The purpose of the table is not to suggest that one risk level is more appropriate than others; instead the table is intended to point out equivalently risky choices once the taxpayer has decided upon an ac-ceptable relative risk level.

For example: Assume our taxpayer decides to accept no risk level in excess of 90%. He is 48 years old with an account value of $500,000 and would like to receive $50,000 per year in distributions. We can solve for the interest rate needed using the “@PMT” function in either Lotus or Excel. In this case, the interest rate is 9.6%. Referring to the table, we can see that the exhaustion percent-ages are between 113.6% and 124.8% for 9% and 10% respectively. Thus, we have a clear signal that the combination of factors (age, assets, distributions, and risk) do not work. To bring down the exhaustion percentages, our taxpayer has a variety of options. He can delay the program a year or two, immediately add to the underlying asset base by including another IRA he had originally held in reserve, or increase his risk tolerance. If our taxpayer waits one year to commence SEPPs (thus aged 49) and his asset base grows to $575,000 in the intervening year, we can again solve for a new interest rate. In this case the interest rate drops to 8.1%. Again referring to the table, a 49-year-old at 8% has an account exhaustion per-centage of 94.9%. Although this is not under 90%, it is certainly closer than before and potentially close enough that the taxpayer is willing to accept 5% additional risk and commence SEPPs.

Another potential solution to asset exhaustion is the concept of “account replenishment.” This is very specifically a concept that has never even been discussed in a private letter ruling. Assume our taxpayer splits his IRA into two IRAs (A and B) before commencing SEPPs. He commences SEPPs on IRA A, holding IRA B in reserve for a later date. Misfortune occurs, and IRA A is 100% depleted be-fore our taxpayer reaches age 59½. Can our taxpayer replenish IRA A with assets from IRA B? No one knows for sure. On the one hand, this type of action would seem to violate our earlier concept of the fixed SEPP universe of accounts. On the other hand, the idea of replenishment would certainly be handy as a course of last resort. Needless to say, anyone facing this precarious a situation would be best advised to seek expert advice, which would almost certainly lead to a filing for a private letter rul-ing.

All of the above have been predicated on two assumptions: 1. Account exhaustion risk exists.

2. There is a mathematical or quantifiable method for risk measurement.

Dealing with these assumptions in reverse, the table presented at the end of this chapter is a way to measure this risk but is not necessarily the only way to measure the risk. Further, the table has no asset growth assumptions. We could have built the same table with 4% or 6% or 8% asset growth rates. Is this any more realistic than 0%? No one plans for 0% growth; however, if we broaden our time horizon

36

and take into account that we are attempting to quantify a down-side risk only, then a 0% growth rate appears to have some rationale83.

The first assumption was that account exhaustion risk exists. In most circumstances this is true, but not in all cases. For starters, those taxpayers electing to use the minimum method need not read about interest rates and account exhaustion at all. Taxpayers electing the amortization or annuity methods are exposed to this risk on the presumption of a one-time calculation of program distributions. Further, this risk actually rises for a one-time calculation coupled with a fixed or variable cost-of-living adjustment each year. Conversely, this risk completely vanishes if the taxpayer elects annual recalculation in con-junction with either the amortization or annuity method. Thus, annual recalculation removes account exhaustion risk. However, there are new risks associated with annual recalculation, which are dis-cussed in a later chapter.

External Risk Management So far, we have discussed the internal risks associated with the selection of an interest rate. Equally important are the external influences on interest rate selection. Here we turn to all existing private let-ter rulings through January 2000 in search of some patterns of approved rates. Between 1988 and 2000, the IRS issued 56 private letter rulings relevant to §72(t)(2)(A)(iv). In 40 of these 56 PLRs, the interest rate requested by the taxpayer is displayed within the text of the PLR. In the other 16 PLRs, the inter-est rate was either not relevant, e.g. a PLR using the minimum method, or the interest rate was re-dacted84 for some reason. Those interest rates used have been compared to the long-term applicable federal rate in effect for the same month as the PLR was issued85. The results of those comparisons are as follows:

• 16 of the PLRs were quoted as relative to either the applicable federal rate (AFR)86 or Pension Benefit Guarantee Corporation (PBGC) rate in effect at that time. These relative quotes were either “on the rate” as quoted or the relevant rate plus or minus X%, where X was never greater than 20%.

• 24 of the PLRs quoted absolute rates ranging from a low of 5% to high of 10.6%. When com-paring these absolute rates to AFRs in force at the time, the following occurred:

• 17 out of the 24 times, the absolute rate was within ±20% of the extant AFR. Of the seven exceptions, five were high and two were low.

• 19 out of the 24 times, the absolute rate was within ±25% of the extant AFR. Of the five exceptions, three were high and two were low.

83 However, it is certainly possible for a taxpayer to invest in an account exhaustion risk-free manner, such as through the purchase of laddered U. S. Treasury securities with maturities matching the SEPP payment dates. Such a program would theoretically elimi-nate all account exhaustion risk such that the taxpayer could solve for the maximum interest rate assumption that is tolerable by leaving a positive account balance in the SEPP account at the end of the required SEPP period. 84 Remember that redacted data is at the request of the taxpayer, not the IRS. Therefore, we no reason to suspect that the IRS has arbitrarily redacted some outrageously high interest rates. 85 Notice 89-25 tells us to measure the interest rate when payments commence. Unfortunately, in many PLRs it is not pos-sible to accurately determine the actual payment commencement date. As a result, we have substituted ruling issuance month on the assumption that a taxpayer receiving a positive letter would commence payments shortly after receipt of the letter. 86 Most often, the long-term applicable federal rate was quoted; however, occasionally the mid-term rate was used also.

37

• 21 out of the 24 times, the absolute rate was within ±30% of the extant AFR. Of the three exceptions, two were high and one was low.

On the presumption that the IRS would more likely challenge a high rate than a low rate, let’s look closer at the high rates. Of the five high exceptions, four were within 1% absolute, and only one was beyond a 1% difference. The one high exception beyond 1% occurred in April 1998, when the long-term AFR was 5.98%. Therefore AFR plus 20% was 7.18%, and this particular taxpayer elected an ab-solute rate of 8.25%, which is 1.07% greater than AFR plus 20%. The average high exception was 0.6% higher than AFR plus 20%.

From the above data, we can conclude the following four points: 1. As a general observation, taxpayers seemed to act conservatively87 when explicitly suggesting an

interest rate within a PLR request. Even the highest absolute rate quoted, 10.6%, was only 0.11% higher than AFR plus 20%, which at that time was 10.49%.

2. In 35 of 40 observations, the interest rate was either relatively stated or within AFR plus 20%. This appears to be statistically significant such that we should consider 120% times AFR as a material upper barrier that should not be lightly breached.

3. 97.5% of the time (there was the one exception that missed by .07%), the explicit interest rate as-sumption in the PLR was never greater than AFR plus 20% of AFR plus 1%.

4. Based on the above, the upper limits become pretty clear. Risk-adverse taxpayers should stay within AFR plus 20%. Risk-prone taxpayers might want to venture past 120% of AFR but certainly not by more than 1%. Somewhere beyond AFR plus 20%, a taxpayer should seriously consider ap-plying for a new PLR to support that higher interest rate.

This author has dealt regularly with the IRS88 on this issue. As of early 2001, the IRS’s position is that it officially (but only orally) sanctioned the mid-term applicable federal rate plus or minus 20%. However, in a manner, the IRS speaks with forked tongue. Its attitude is that SEPPs should be engaged with a long-term or lifetime view (even though SEPPs are required to run only for the greater of five years or attaining age 59½). This would then suggest that the correct rate to use is the long-term appli-cable federal rate more closely corresponding to the concept of a lifetime. As a result, without too much arm twisting, the IRS regularly approves the use of the long-term applicable federal rate plus or minus 20%.

Conclusions Selecting an interest rate is not a simple task. Our taxpayer needs to consider internal issues of ac-count exhaustion risk as well as external issues that focus on IRS approved rates. Internal risk is manageable by either:

• Investing in an underlying risk-free portfolio that eliminates the possibility of account exhaus-tion.

87 Unfortunately, there does not seem to be any viable data collection technique to determine what interest rates are being used by the thousands of taxpayers who are taking SEPPs but do not have private letter rulings. 88 In this particular instance, it is the Assistant General Counsel’s Office -- Employee Plans and Exempt Organizations -- that is responsible for private letter rulings on §72(t) matters.

38

• Switching to annual recalculation using updated account balances.

• Subjectively managing the age/interest rate to account exhaustion relationship.

• Using account replenishment techniques as a last attempt to avoid account exhaustion with the likely requirement of filing for a private letter ruling.

External risk is manageable by either:

• Selecting an interest rate that is equal to or less than the extant applicable federal rate times 1.2.

• Filing for an individual private letter ruling for more aggressive interest rates.

Table of Account Exhaustion Percentages Based on Age and Interest Rate Age 4.00% 5.00% 6.00% 7.00% 8.00% 9.00% 10.00% 11.00% 12.00% 40 98.62% 114.38% 131.01% 148.37% 166.32% 184.74% 203.54% 222.64% 241.96%

41 94.57% 109.45% 125.15% 141.54% 158.50% 175.92% 193.71% 211.79% 230.09%

42 90.39% 104.40% 119.19% 134.63% 150.62% 167.05% 183.84% 200.90% 218.19%

43 86.25% 99.40% 113.27% 127.77% 142.78% 158.21% 173.99% 190.05% 206.32%

44 81.97% 94.27% 107.25% 120.81% 134.86% 149.32% 164.10% 179.16% 194.42%

45 77.72% 89.17% 101.26% 113.89% 126.98% 140.45% 154.24% 168.29% 182.55%

46 73.31% 83.94% 95.15% 106.86% 119.01% 131.52% 144.33% 157.38% 170.64%

47 68.84% 78.65% 88.99% 99.79% 111.01% 122.56% 134.39% 146.46% 158.71%

48 64.38% 73.37% 82.84% 92.75% 103.02% 113.61% 124.47% 135.55% 146.81%

49 59.75% 67.93% 76.56% 85.58% 94.93% 104.58% 114.48% 124.58% 134.86%

50 55.02% 62.41% 70.20% 78.34% 86.79% 95.51% 104.45% 113.59% 122.89%

51 50.20% 56.81% 63.77% 71.04% 78.59% 86.39% 94.39% 102.56% 110.88%

52 45.26% 51.10% 57.24% 63.66% 70.32% 77.20% 84.27% 91.49% 98.85%

53 40.20% 45.27% 50.61% 56.18% 61.97% 67.95% 74.09% 80.37% 86.77%

54 35.01% 39.32% 43.86% 48.60% 53.53% 58.61% 63.84% 69.18% 74.64%

55 29.66% 33.23% 36.99% 40.91% 44.98% 49.18% 53.50% 57.93% 62.44%

56 24.15% 26.99% 29.97% 33.08% 36.31% 39.64% 43.07% 46.59% 50.17%

57 18.45% 20.56% 22.78% 25.09% 27.50% 29.98% 32.53% 35.14% 37.81%

58 12.54% 13.94% 15.41% 16.94% 18.52% 20.16% 21.85% 23.58% 25.35%

59 6.40% 7.10% 7.82% 8.58% 9.37% 10.18% 11.02% 11.87% 12.75%

39

_______________________________________________________________________________________________ Chapter 6 Life Expectancy Table Selection Approved Mortality Tables Of the three currently approved methods, only the annuity method affords the taxpayer the flexibil-ity of selecting from a variety of different life expectancy tables. Notice 89-25 uses the example of the UP-198489 as have we in previous chapters. UP-1984 is convenient, as it is readily available from a variety of sources in both written and electronic format. In addition, the Secretary of Treasury is authorized to approve a variety of mortality tables accord-ing to IRC §1.401(a)(4)-12. The tables currently approved in addition to UP-1984 include:

• 1983 Group Annuity Mortality Table (1983 GAM) (Female)

• 1983 Group Annuity Mortality Table (1983 GAM) (Male)

• 1983 Individual Annuity Mortality Table (1983 IAM) (Female)

• 1983 Individual Annuity Mortality Table (1983 IAM) (Male)

• 1971 Group Annuity Mortality Table (1971 GAM) (Male)

• 1971 Group Annuity Mortality Table (1971 GAM) (Female)

• 1971 Individual Annuity Mortality Table (1971 IAM) (Female)

• 1971 Individual Annuity Mortality Table (1971 IAM) (Male)

These tables are also available for a modest fee from the Society of Actuaries. However, before anyone runs out to purchase any of these tables, we should examine what a mortality table is and how one converts a mortality table to an annuitant calculator or annuitant divisor table.

For example: Let’s use UP-1984. Looking in Appendix G at the age-50 row, we can see that the 50-year-olds started with a beginning population of 72,180,169.

% Deaths Life

Age Per Age Population Deaths Expectancy 50 0.5616% 72,180,169 405,364 26.89 51 0.6196% 71,774,806 444,717 26.04

One year later, we can see that 405,364 of the 50-year-olds died before reaching their 51st birth-days. Similarly we can say that 0.56% of the 50-year-olds expired, and 99.44% made it to age 51. If we contrast a 25-year-old with a 50-year-old, we can see that the 25-year-old has a life expec-tancy of 50.02 additional years. Said another way, a 25-year-old can, on average, expect to attain the age of 75.02. The 50-year-old has a life expectancy of 26.89 or an average age of 76.89. How is it that the 50-year-old has a greater life expectancy than a 25-year-old? Simple, life expectancy is a

89 The UP-1984 mortality table is published by the Society of Actuaries, 475 North Martingale Road, Suite 800, Schaum-burg, Illinois 60173. The table, only, has been reproduced in Appendix G.

40

function of one’s attained age. Said another way, the 50-year-old has already visited his 26th through 50th birthdays and made it successfully, whereas the 25-year-old has not. Jumping forward another 25 years; if a woman makes it to her 75th birthday; then, on average she will live another 9.49 years attaining an average age of 84.49.

So how do we convert any mortality table into an annuitant divisor table or annuitant table? It is actually not as difficult as it might appear. All we need is a working copy of Lotus, Excel, or any other spreadsheet product. Annuitant Tables90 are age and interest-rate sensitive. To demonstrate arriving at an annuitant divisor table, we will return to John Q. Taxpayer, age 52, and an assumed interest rate of 8%. The column layout is as follows:

ColA ColB ColC ColD ColE ColF ColG ColH ColI Published UP1984 Probability of UP1984 Raw Computed Living After NPV of $1 Annuitant Percent To

Age Factor Mortality (1) Factor Difference Age Attain. @ X% Factor Original

51 0.006196 71774805 0.006196 0.0000 0.0000 0.0000 0.0000 71.77% 52 0.006853 71330089 0.006853 0.0000 1.0000 1.0000 1.0000 71.33% 53 0.007543 70841264 0.007543 0.0000 0.9931 0.9259 0.9196 70.84% 54 0.008278 70306908 0.008278 0.0000 0.9857 0.8573 0.8450 70.31% 25.7004 13.3667 10.8452

Please refer to Appendices G and H for the complete tables.

Col A: Ages 15 to 111

Col B: Published % of deaths per age (see Appendix G) Col C: Raw Mortality (also from Appendix G)

Col D: Computed Factor: [(Col C, Row X) – (Col C, Row X+1)] / (Col C, Row X)91 Col E: Difference (Col B minus Col D)92

Col F: Is the probability of continuing to live after attaining one’s current age. This can be ex-pressed algebraically as “IF Col A = 52, Put a 1 here, Otherwise, IF Col A < 52, Put a zero here, ELSE put a number here equal to (1 - Col B) * (Col F, up one relative row)”.

What’s happening here? We want this column to be a zero for all ages before 52 because John has already passed through those ages. Further if John is age 52, then he has a 100% probability of attaining that age because he is that age, thus the “1.” For any age above 52, John has a probability of living to that age that is less than 100% that is equal to (1-

90 You may wish to refer to Appendix H which is formatted to match the following example. 91 As an example, for a 52 year-old, the math is: (71330089 – 70841264) / 71330089 = 0.006853. 92 Actually, the only columns we really need are A and B. We created columns C, D, and E as a simple proof to test that the factors keyed in column B are correct.

41

Col B) the percent of deaths for that age times the probability of attaining the immediately previous age. Appendix H shows a “1” in this column for age 52. John has a 99.3147% probability of making it to age 53, a 98.5656% probability of making it to age 54, and so forth.

Col G: Net Present Value of $1 At the Chosen Interest Rate. This also can be expressed as: IF Col A = 52, put a 1 here.93

IF Col A < 52, put a 0 here. IF Col A > 52, put a value here of (1 / (1 + .08) * the value in Col G up one row).

Essentially, because John is age 52, the value of $1 is $1 this year. However, the value of a $1 one year from now is worth less than $1. It is worth only $1/(1.08) or approximately 93½ cents. Two years from now, $1 is worth only $1/(1.08)*(1.08) or approximately 85½ cents, and so forth.

Col H: Is the Annuitant Factor for one year, expressed as: Col F times Col G The sum of Col H is the annuitant divisor, which in this case is 10.845. As John ages,

let’s say two more years to age 54, the divisor will get marginally smaller to 10.562. Fur-ther, if we increase the interest rate from 8% to 12%, the divisor will also get proportion-ately smaller, in this case to 8.264. Since the beginning balance of the IRA is divided by the annuitant divisor, the smaller a divisor the larger the computed annual withdrawal. So as ages and/or interest rates rise, so will the annual distribution. Similarly as ages and/or interest rates fall, so will the annual distribution.

So, which table, given that there are currently nine to pick from, should a taxpayer use in conjunc-tion with annuity method? Always use UP-1984 for several reasons:

1. It is readily available, fixed (it has already been around 17 years and will not change), and is formally and directly IRS approved in a standing regulation.

2. More importantly, UP-1984 is the most aggressive of the nine approved mortality tables, ag-gressiveness being defined as the table with the earliest and cumulative most frequent deaths. This results in the lowest annuitant factors of all of the tables at the same age/interest rate inter-sect. The lowest annuitant factor results in the highest annual distribution in SEPPs holding age and interest rate constant.94 Therefore, there is no need to build annuitant divisor tables for any of the other approved mortality tables.

Knowing how an annuitant table functions, we can more easily see the flexibility afforded us in the annuity method. John can’t change his age, but he can certainly change interest rates (as discussed in the preceding chapter), and he can pick from nine different pre-approved mortality tables. As a result, with a little pre-planning of IRA balances,95 interest-rate assumptions, and mortality table used, John can develop a literal continuum of annual withdrawals. Because the annuity method provides the

93 Those individuals who intend on performing multiple iterations, e.g. modeling of results, will be better served by making the formulas in columns F and G conditional based on age and interest rate variables housed elsewhere in the spreadsheet. 94 If a taxpayer derives an annual distribution amount that he considers too high for the circumstances or needs, it is always easy to reduce the distribution amount by account fracturing or reducing the interest-rate assumption. 95 We will talk extensively about IRA fracturing and aggregation in the next chapter. In this example, it is fair to assume that John can select just about any IRA opening balance he wants between $100,000 and $1,000,000.

42

maximum flexibility of all three methods, many taxpayers chose this method and implement it with some backwards-thinking logic. The question is not, “What will the annual withdrawal be for John, aged 52, using 8% and UP-1984?” Instead, the real question is, “What annual withdrawal amount does John really want?” Then structure the opening IRA balances, the interest rate, and potentially the mor-tality table to cause the mathematics to compute correctly.

For example: John can assume a 5% interest rate and withdraw approximately $7,100 per year per $100,000 of beginning assets. At the other end of the spectrum, John can assume 10% and with-draw approximately $10,700 per year, 50% more than at 5%.

There is a practical limit to this exercise but only on the up side. Any smaller annual withdrawal is always achievable, simply by reducing the opening balance of the IRA, reducing the interest-rate as-sumption, or moving back to either the amortization or minimum methods. The upside, (meaning the largest annual withdrawal amount) is limited by using the highest interest-rate factor reasonable (as discussed in the preceding chapter) coupled with the most aggressive of the nine mortality tables.96 UP-1984 is a blended unisex table that comprised 80% males and 20% females. To quote directly from the study summary, “A one year set-forward in the age in the basic tables produces mortality rates that may be considered to represent the pensioner mortality experience of an all male group on a fairly accurate basis.… Similarly, a four-year setback in age in the basic UP-1984 Table results in a set of mortality rates that may be considered appropriate for use in valuing pensions for an all female group.” The above was written for actuaries at insurance companies, which basically suggests that to convert the UP-1984 Table from a unisex table into male and female tables one need only add one to the age of a male and subtract four from the age of a female. Well, if that is the case, when we use UP-1984 to build an annuitant divisor table, should we add one and subtract four respectively? The effect will be to slightly raise distribution amounts for males by approximately 1.5% and lower distribution amounts for females by approximately 4.5%. As usual, the answer is unknown—simply because no one has asked the IRS for a direct answer.97 Anyone contemplating early retirement should always have an annuitant divisor calculator at his fingertips, even if he does not intend on using the annuity method. The reason is that the annuitant di-visor calculator is always quick and provides a one-minute upper boundary threshold98. Recognizing that the mathematics for building your own divisor table (described preceding) might appear a bit daunting, we suggest you take a short-cut to: http://www.retireearlyhomepage.com/. At this web site there are several free and downloadable annuitant divisor tables. Whether you build your own from scratch or download a table from elsewhere, the mathematics should always be double checked99 for complete accuracy. Finally, it is always helpful to have real interest rates at hand. This author uses: 96 The most aggressive mortality table is the one that, depending on sex, causes the most deaths at the earliest ages. 97 This author suspects that it would be appropriate to add one and subtract four; no one, however, has asked the direct question most likely because it was never worth it to ask. Anyone desiring to do so would be well advised to make a private letter ruling submission since this issue has never been tested. 98 As an example, if you are 50 years old, have a $1,000,000 IRA and 120% of the long-term applicable federal rate is 8%, the annuitant divisor is 11.109; therefore, in seconds one can determine that the upper threshold of annual distributions is approximately $90,000. Any distribution of less than $90,000 is achievable using all the other tools we have discussed. 99 It would be really embarrassing, as well as costly, if one built (or downloaded) a mathematically incorrect table and there-fore miscalculated an annuity payment amount. In addition to cross-checking against Appendix H; here are some further data points at 10%: age 40 – 10.255; age 45 – 9.956; age 50 – 9.071; age 55 – 9.071. If your table does not cross-check to these values, your table has either a data input or logic error.

43

http://www.timevalue.com/afrindex.htm in that it is an easy to use web site with 110% and 120% of applica-ble federal rates pre-calculated. Of course, taxpayers can always go to http://www.irs.gov and look up the correct Revenue Bulletins, which contain the monthly updates to applicable federal rates.

Table S: 80CNMST & 90CM Table S is an IRS published annuitant divisor table currently based on the 90CM100, an IRS pub-lished mortality table101 based on the mortality data from the 1990 census. Most of IRS Publication 1457 is devoted to the computation of annuitant divisor values for people ages 0 to 109 in one-tenth of a percent increments from 4% to 25%.102 Some have suggested that the 90CM table can be used in lieu of UP-1984 or other authorized tables. The advantage of doing so would be that 90CM is a little more aggressive than UP-1984. As an example, John at age 52 assuming 8% gets an annuitant divisor from UP-1984 of 10.845. Without aging a day, John could switch to 90CM and get an annuitant divisor of 10.2256, approximately a 5.75% decrease in the divisor and a corresponding increase in the annual withdrawal amount.

Thus, a taxpayer is really left with three choices: Use UP-1984 as a unisex table, add one (or sub-tract four) to the age and use UP-1984 (for males only), or use 90CM. This author suggests that 90CM should be used only judiciously, meaning through use of a private letter ruling, for two reasons. One, it is not one of the listed tables approved by the Secretary103; two, it is the AGC’s office present position that Table S inappropriate for use in the computation of the annuity method. Rather, their position is that Table S’s purpose in life is the computation of remainderman interests104, most often as a result of charitable gift giving; conversely, §72(t) annuity computations are most appropriately done using any one of the nine tables approved105.

100 90CM mortality table and the resultant update to Table S in Publication 1457 was just published in the Fall of 1999; replacing the 80CNMST table, which had been in force for the preceding 10 years. The IRS is statutorily required to repub-lish Publication 1457 no less frequently than dec-annually using the then available most recent census data. Thus, it is rea-sonable to presume that we can expect to see a “00CMT” table sometime in 2009. See IRC §7520(c)(3). 101 Publication 1457: Actuarial Values: Book Aleph is available from the Government Printing Office for $65 and also usu-ally available at college or public libraries that act as federal government document repositories. 102 This creates a 600-page-plus document. Actually, if one thinks about it, paying $65 for a 600-page document is rather rational. 103 So far, there is one PLR (#98-24047) that approved the use of 80CNMST for one taxpayer. The IRS’s unofficial position is that 80CNMST and now 90CM are inappropriate for use in conjunction with §72(t). 104 Some of might wonder what is a remainderman interest? Assume you are wealthy and give a $1,000,000 bond portfo-lio to the charity of your choice in return receiving a life estate annuity of $60,000 per year. Further, lets assume an interest rate of 8% and a life expectancy of 30 years. This makes the annuity payments to you worth approximately $675,000; therefore the charity is really receiving a “remainderman interest” donation of approximately $325,000; not $1,000,000. This area of the tax code was often abused through the manipulation of interest rate assumptions in the 1980’s and 1990’s until the IRS forced uniformity in mortality and interest rate usage. 105 Further, IRC §7520(b) specifically says in part, “this section shall not apply for purposes of part I of subchapter D of chapter 1….” Well, part I of subchapter D of chapter 1 of the IRC is none other than §§401-424. §72(t) is linked to §§401-424 (via an intermediate definitional stop @ §4974(c)). Therefore, it is pretty easy to come to the conclusion that §§401-424 and therefore §72 are either, at a minimum, excused, or at a maximum, forbidden from using Table S. This is actually appropriate; otherwise §§401-424 and §7520 would be in direct conflict with one another.

44

All this being said, there are taxpayers out there today that have based their annuity method calcu-lations on the either Table S circa 1990-1999 using the 80CNMST mortality table or the Table S circa 1999-present using the 90CM mortality table. What should these taxpayers do? Unfortunately, it is a mess:

• If you properly used either version of Table S as a one-time annuity method calculation, stick with it; there is no reason to change.

• If you improperly used Table S106 as a one-time annuity method calculation; e.g. you computed your distribution using 6/31/99 balances (or later) for a calendar 1999 or a 2000 distribution and used the old Table S107 your computation is incorrect and some corrective measures need to be taken.

• If you are using Table S as your annuitant table and are performing annual recalculation, you should have “stepped-in-stride” switched from old Table S to new Table S in either 1999 or 2000, depending on what month you perform your annual recalculation.

What should you do in the second circumstance above or if you didn’t step-in-stride as you were supposed to in the third circumstance? Unfortunately, this author doesn’t know. On one hand, igno-rance of the law is not a defense; on the other hand, going back into prior years and changing the calculations or the distributions seems virtually impossible. In short, this issue has not come up in either a PLR or in litigation but does appear to be a catch-22, where there is no apparent mechanism to repair the situation. There are several positive sides to the problem in that, one, the Table Ss are fairly well hidden from general public view (the update, even more so) and, two, the issuance of the new Table S was of IRS manufacture as opposed to having been directly contained with the IRC. Potentially, the IRS may consider some form of leniency should your Social Security number win the audit lottery. The only other potential solution would be to carefully craft a private letter ruling submission in an attempt to gain some form of relief.

106 The IRS, in its infinite wisdom, has decided that old Table S using 80CNMST was valid through April 30, 1999, and conditionally valid (meaning optional) through June 30, 1999; the new Table S using 90CM was conditionally valid for April 30, 1999, through June 30, 1999, and came into full, mandatory effect on July 1, 1999, and forward. See Treasury Decision 8819, 64 FR 23187-23229, April 30, 1999. 107 There is a fairly significant problem here in that most taxpayers would not know that Table S existed unless they hap-pened upon it in some other text. Even fewer taxpayers know that there are two versions of Table S much less know which one is which and which is the correct one to use. Nonetheless, if you are in this predicament, closely examine whatever Table S you are looking at. At the top, it should say somewhere that it is using 80CNMST or 90CM. If you can’t tell, throw away the table.

45

_______________________________________________________________________________________________

Chapter 7 Other Planning Issues Some one once said success or winning is in the details. In our opinion, this is particularly true in the computation of SEPPs. Thus far we have covered general concepts, method computation specifics, interest rate selection, and life expectancy tables. Now, we need to put all of this together to effectively plan a SEPP program. But, before we can do that, there are a variety of tactical planning issues that require some coverage. What follows are planning tips. Sometimes these tips are obvious and simply need repeating within the context of a SEPP.

Usually, these tips are founded in other sections of the IRC or related documents of authority. Other times, these planning tips are the author’s opinion combining authority and patterns in the PLRs with in-depth experience on this subject.

How to Plan a SEPP This author’s opinion is that SEPPs should be planned in reverse. Step one should be to quickly use the annuity method with a modestly aggressive interest rate based on the sum of all of your deferred accounts. This is a “back of the envelope” quick calculation to determine the upper limit. Using John again, at age 52, 8% and $1,000,000 in IRAs—the annuity method tells us that John’s upper limit is in the $90,000 to $92,000 range. John may have already decided that he needs $120,000 per year for the next eight years. Maybe by further pushing of the envelope, we can increase his SEPPs to $95,000 or $98,000, but we cannot get them to $120,000. John should stop right here and reassess his living needs or delay program implementation.

As a second example, let’s assume John has decided that he needs $60,000 per year for the first four years and $80,000 per year for years five through eight (to cover, say college tuition for John, Jr.). Now we have something to work with—a set of cash flow expectations that are under the upper limit. For the moment, let’s forget about methods, interest rates, and so forth. Instead, we should put all of our energies into mapping out cash flow needs in as much detail as possible. This implies some budgeting for both regular living needs, capital expenditures, maybe tuition or elder care, as well as some kind of unexpected emergency needs. Once we have cash flow figured out, planning SEPPs is really rather easy. We make SEPPs fit the cash flow needs, not the other way around. What follows are simply a set of tactical formation tools that will allow you to custom tailor a SEPP program or series of SEPP programs to fit your needs.

Account Fracturing and Aggregation All of our previous examples have focused on a single IRA with a single balance. This is fine for example purposes, but it poorly reflects real life. More often than not, a taxpayer will have a collection of deferred accounts with wide-ranging values. The IRC is very clear on the subject of account fractur-ing (one into many) and account aggregation (many into one). As long as the IRAs are moving be-tween or within approved trustees, anything is game. John can have four IRAs or forty (by fracturing) or twenty-one by combining the last twenty into one (aggregation). How do we interpret or use these rules in light of our desire to create one or more SEPP programs?

46

First, all fracturing108 or aggregation relative to the accounts to be used for the SEPP program must be completed before the first SEPP withdrawal. This defines the account universe from which SEPPs will be performed. Second, once the first SEPP has physically occurred, the account universe (which can be one or several deferred accounts) is cast in concrete for the duration of the SEPP program. Accounts within the universe must stay inside the universe and cannot leave. Similarly, accounts outside the universe cannot enter. Another way to think about this is that the only cash transaction permitted within the ac-count universe is the periodic withdrawal transaction, which is the SEPP. To add an account after the first SEPP transaction looks like replenishment; to remove an account almost looks like a withdrawal of sorts, thus a modification.

Third, any fracturing or aggregation on other non-SEPP accounts can be left for the future. Fourth, one must strike a balance between committing assets to the account universe in order to receive SEPPs and holding other assets outside the account universe (housed in separate IRAs) for fu-ture needs and/or risk exposure.

Multiple SEPP Programs109 Why not? An excellent tactical move for John is to commence SEPP program #1 immediately to provide his $60,000 per year in living needs and commence SEPP program #2 four years in the future to cover the college tuition payments. What are the rules to make this work effectively?

1. The account universes for program #1 and program #2 are discrete and cannot overlap 2. SEPP program #2 commences at a different date and may use completely different assumptions

than program #1. As an example, program #1 can use the annuity method, John aged 52 and 6% interest. Program #2 can use the amortization method, John aged 56 and a 7% interest rate assumption.

3. SEPP withdrawal dollars for program #1 must be withdrawn from an account(s) that is within universe #1. Similarly, SEPP withdrawal dollars for program #2 must be withdrawn from an account(s) that is within universe #2.

Important Dates There are a variety of calendar dates that are important:

• Your birthday

• SEPP starting date

• Account valuation dates

108 In one of the few adverse PLRs, in 97-05033 the Service ruled: “the entire account balance in each of the pertinent IRAs must be taken into account. That is, a portion of one or more of the IRAs may not be excluded in order to limit the periodic payment to a predetermined amount.” The ruling itself is very clear; what is unclear is whether the taxpayer had intended to fracture one of his IRAs before commencing his SEPPs and simply neglected to do so, or whether his independent advisor simply made an interpretive or math error. Nonetheless, the ruling is clear; get all of you account fracturing done before the first withdrawal. 109 See PLR 98-12038.

47

Each of these dates is important for different reasons. Further, these dates actually have little rela-tive importance in computing the dollar value of SEPPs, but they are administratively critical, e.g. date errors are easy to make and have a small, but real, possibility of invalidating a SEPP program. Unless otherwise specifically noted, IRC §§401-416 (and therefore §72) always uses the concept of “highest attained age within the tax year.” As an example, if your birthday is 6/30/50 and it is now 6/29/00, you are 49. One day later you are 50. Nonetheless, if you commence SEPPs in 3/00; you are 50, because 50 is your highest attained age within tax year 2000. SEPPs must continue for at least five years. When do we start counting, and how do we count? A SEPP program’s start date is the calendar date of the first withdrawal from one or more of the accounts defined in the SEPP universe. From that date, we literally count 1828 days.110 1828 is 365 times 5 plus the potential for two leap years; plus one day for insurance. If the first SEPP cash withdrawal occurred on 3/15/00, the easier way is to say that the five years expires on 3/16/05.

Accounts can be valued on any date. Just as an administrative matter of proving a valuation, this author always suggests that a month-end be used. If necessary, going forward, accounts should always be revalued on the same date one year later. SEPPs must continue until you are age 59½. When are you 59½? Exactly 183 days after your 59th birthday. If your birthday is 6/30/50, your 59½ birthday is 12/30/09. In order to completely satisfy the §72(t) rules, SEPPs must continue for at least five years and until you are age 59½. You know the SEPP start date; write it down. You can easily compute SEPP start date plus five years as well as compute your 59½ birthday. Use the later of these two dates and discard the earlier. When executing SEPP withdrawal transactions, simply insure that only SEPP transactions occur between these two dates.111

Here’s another way to compute SEPP ending dates: If you started your SEPP on or before your 54½ birthday, your earliest ending date is always your 59½ birthday; conversely, if you started your SEPP after your 54 ½ birthday, your earliest end date is always the SEPP start date plus 1828 days.

Stub Periods What’s a “stub” period? Any period of time that is less in duration than one year. Suppose John was employed through May 2000 and would like to start SEPPs of $60,000 per year in June but does not need that much in 2000 because he had five months of earned income from his prior employer. Must John wait until 2001 to start SEPPs? Absolutely not. John can define the program in any of three ways: 1) an annual program of $60,000 per annum; 2) a quarterly program of $15,000 per quarter; or 3) a monthly program of $5,000 per month.112 As a result, John could:

• Take a full $60,000 in 2000 under the annual concept.

110 Remember the Arnold v. Commissioner case (111 TC 250; 1998 U. S. Tax Ct.) In this case, the Court sided with the IRS in a literal interpretation of the IRC that five years was indeed the effective equivalent of 1828 days. Although it has never been litigated, the same question would apply to determining when a taxpayer turns 59½. Because we are discussing the interpretation of the IRC, as opposed to an IRS announcement of some kind, it is the author’s opinion that the safest inter-pretation is a literal one: adding 183 days to one’s 59th birthday. 111 One additional non-SEPP withdrawal made one day too early would technically be considered a modification to the SEPP program potentially causing the application of the 10% surtax. 112 See PLRs 90-10075, 90-31045, 90-46064, and 90-49044.

48

• Take either $30,000 or $45,000 in 2000 under the quarterly concept.

• Take $35,000 under the monthly concept for the months of June through December. Whichever concept John chooses is fine, but it applies only in the first year. Commencing in 2001; John must take a full $60,000.

Disbursement Frequencies and Locations We can define the account universe (implying two or more physical accounts) from which cash disbursements will be made in order to satisfy the SEPP program. Let’s assume two accounts, but it could just as easily be 10. When you decide to make a SEPP disbursement, that disbursement can come from any one or more than one account that was originally defined as a member of the account universe. As an example, if the universe comprises two accounts and you wish to make a withdrawal of $10,000, you may withdraw $10,000 from account one, $10,000 from account two, $4,000 from ac-count one and $6,000 from account two, or any other permutation of dollars that suits your needs. Frequency of SEPP distributions is equally flexible. Assuming the annual amount is $36,000, you may withdraw $36,000 all at once, $3,000 a month, $10 per day for the first 360 days, or any other random permutation of withdrawals that meets your needs as long as the sum total of all withdrawals equals exactly $36,000.

Diminimus Issues “Diminimus” equals small or insignificant on either a relative or absolute scale. What is diminimus to the Internal Revenue Service? One dollar. The IRS permits rounding off of pennies on a tax return, but you can’t round to the nearest $10 or $100. This concept actually comes into play in two circum-stances—calculations and cash transactions.

Calculation—John, aged 52 with $1,000,000 in his IRA chooses to use the minimum method. The calculation is $1,000,000 / 31.3, which equals exactly $31,948.88. Therefore $31,948 and $31,949 are both acceptable as diminimus penny dropping or rounding respectively. $31,950 is not acceptable. Transaction—John needs to withdraw $31,949. He makes a series of withdrawals of the correct period of time that total exactly $31,948.50. This difference will also be treated as diminimus. Con-versely, if John accidentally withdraws $31,952, the difference will not be considered diminimus.

Any nondiminimus difference has the potential to be treated as a modification resulting in the im-position of the 10% surtax. In this author’s opinion, if John missed by $3 or even $10, the IRS would most likely not call it a modification. However, why take the chance, particularly when there is nothing but downside risk. Calculate, recalculate, and stick to the correct amounts within one dollar.

Cost of Living Adjustments The minimum method has a built in cost-of-living adjustment of sorts. Each year, the minimum method is usually implemented with annual recalculation using an updated account balance (hopefully higher) and an updated life expectancy (that will be smaller). The result is an increased withdrawal amount each year.

Conversely (but for the “recalculation concept” discussed in the next section), the amortization and annuity methods are one-time and one-amount determinations. Through a variety of private letter rul-

49

ings,113 the IRS has determined that adding a cost-of-living increase to the fixed amounts as deter-mined by the amortization or annuity methods does represent “substantially equal.”

Cost-of-living adjustments can be fixed or relative. Fixed adjustments are expressed as a percent-age increase, e.g. 2%, 3%, or 4% per year.114 Relative adjustments require the comparative use of some acceptable indexing mechanism; e.g. 12/99 consumer price index - national urban115 divided by the 12/98 consumer price index - national urban. Such a comparison might result in a number like 1.0245. The relative adjustment would then be the prior year’s distribution amount multiplied by 1.0245.116

Recalculation in Conjunction with the Amortization or Annuity Methods This is one of the newest topics in regard to SEPPs and is easily the most difficult topic to discuss relative to SEPPs. We will try to dissect the issue into three pieces: What is it? Do we think it is right? Does the IRS think it is right? When we look at the amortization and annuity methods, there are three variables at work: account balance, age,117 and interest rate. Most often a taxpayer picks a date, let’s say as of some year-end, to make all three determinations: What is the account balance at 12/31/XX; what is my attained age at 12/31/XX; and what interest-rate assumption do I want to make as of 12/31/XX. Drop these three variables into the amortization or an-nuity hopper and out comes a number. Up to this point, we have all assumed that this process occurred once per SEPP program. What is it? Recalculation is the repetition of this process during the SEPP period, presumably on an annual basis using the same methodology but with new variables. The taxpayer cannot switch be-tween the amortization and annuity methods.118 Within one or the other of the methods, however, there seems to exist a window of opportunity to recalculate using some or all updated variables. Thus, we can further narrow the definition of recalculation to mean the mathematical recomputation using either the amortization or annuity methods with an updated account balance, revised interest rate, and up-dated attained age. Theoretically, recomputation may include updating some but not all variables, e.g. holding interest rate constant but updating the account balance and the age, or holding both the interest rate and age constant but updating the account balance, and so forth.

Do we think it is right? We can attack or answer this question on several different levels. First, let’s take an intuitive approach. We all tend to think of home mortgages (the essence of the amortiza-tion method) and immediate, lump-sum annuity purchases as fixed, one-time events that do not involve 113 See PLRs 90-47043, 95-36031, 97-26035, and 98-16028. 114 2%, 3%, and 4% appear to be pretty safe, as they have been accepted multiple times. Anything beyond 4% should be considered risky and would warrant a private letter ruling. 115 Whenever you might see the CPI broadcast on general or financial news, they are invariably referring to the “Consumer Price Index – National Urban” statistic as computed by the U.S. Government. 116 In the author’s opinion, relative COLAs make imminent more sense than a fixed COLA of some percentage. Neverthe-less, a relative COLA has only been positively ruled on once. Therefore, in an abundance of caution, any taxpayer wanting to use a relative COLA is best advised to apply for a separate private letter ruling. 117 Actually, when using the amortization method there are three variables as indicated in the text. When using the annuity method, there is technically a fourth variable—that being the selection of a mortality table. However, since the context at this point is one of recalculation, it is this author’s opinion that recalculation that included a change in the mortality table selected would clearly represent a methodology change and is therefore tantamount to a modification. 118 Switching methods is also intuitively a method change, which would be viewed as modification in the eyes of the IRS.

50

mathematical recalculation at later dates. However, is our own intuition correct? Probably most of the time, but not necessarily all of the time. We now have variable-rate mortgages and lump-sum immedi-ate annuities with variable investment features imbedded within them. Thus, in today’s financial mar-kets, we have products for sale that either require a future period recomputation or have a future period variability in dollar outcomes based on some experience results. Let’s end the intuitive discussion by calling it a draw.

Next, let’s take a legal perspective. Unfortunately, the only place to look is at the detail language imbedded in Notice 89-25. The description of the amortization method includes the phrase “by distrib-uting $8,679 annually.” “Annually” is the plural of “annual.” This naturally suggests that $8,679 is supposed to be distributed for multiple years as opposed to saying something such as, “distributing $8,679 in the first year only with potentially differing amounts in future years.” Further, when we look at the description of the annuity method, we find the language “the taxpayer’s age attained in the first distribution year and continuing for the life of the taxpayer.” This language would also seem to indi-cate that we fix things once; at a minimum, we fix the taxpayer’s age only one time. However, we can just as easily back up one level and say that we are “data mining” the words to the point where the English language (even though it is a crystal-clear IRS document) can fail us. As a result, let’s close the legal perspective and call it a draw as well. Next, let’s consider a computational perspective. If we hold the beginning account balance and in-terest rate constant for the SEPP period, the only variable left is attained age. By increasing one’s age in either the amortization or annuity methods, it results in regularized and modestly higher annual dis-tributions in subsequent years. In a way, it looks like it might have the same computational effect as a cost-of-living increase. This sounds like a positive vote. Conversely, if we are also allowed to update the account balance,119 we can easily get wildly different withdrawal amounts per year depending on the investment results obtained within SEPP accounts. This sounds like a negative vote. Looks like the computational perspective is a draw as well. Does the IRS think it is right? In the end, we could probably devise two or three more evaluative criteria and come up with equally ambiguous and confusing answers. So, from the outside looking in, we have a draw. What’s really important is what the IRS has had to say on this issue. The good news is that the IRS has spoken multiple times. The bad news is that the answers were initially somewhat con-fusing and are mostly contained within those problematic PLRs. Further, there appear to be 12 private letter rulings that tackle this recalculation issue, in whole or in part, from varying perspectives. Unfor-tunately, the vote is 7 - 4 - 1. With that said, let’s jump in and see what we can learn.

The seven PLRs issued in favor120 of recalculation have some common characteristics as follows:

• Across a mix of seven PLRs, all relevant variables were adjusted or updated in subsequent years, although some of the older 1980s PLRs chose to hold one or more variables constant.

• All seven PLRs contained “positive like” language such as calculate, recalculate, determine, compute, etc.

• All seven PLRs were prospective121 in nature, e.g. the taxpayer proposes to do X in the current 119 Let’s face it, the only real reason to recalculate is to use an updated account balance, which will undoubtedly be much larger due to one’s superior investing skills. Updating one’s attained age or relative interest rate just doesn’t account for much of a real difference in dollars. 120 See PLRs 89-11071, 89-19072, 95-31039, 98-05023, 00-51052, 01-05066 and 01-06039. 121 This author believes that the most important feature of winning the recalculation issue is this perspective of “prospec-tive” versus “retrospective.”

51

year and apply X consistently in all required future years. The four PLRs that ruled against recalculation122 also had a variety of similarities including:

• All four PLRs either contained or implied negative language, such as miscalculate(d), modify, retroactive, catch-up, etc.

• All four PLRs were retrospective in nature, e.g. the taxpayer has been doing X for Y years; now the taxpayer wants to go back in time and reset the rules and apply the new rules for all past years as well as all future years.

Although somewhat disguised at first, the one tie123 may actually become a seminal PLR on this issue and, in this author’s opinion, is actually the biggest winner of all. In this case, the taxpayer had a SEPP program in place since early 1993 up through 1998. Come 1998, the taxpayer is either 49 or 50 and has satisfied the five-year rule but has not attained age 59½. The key language is:

Actually, this is a wonderful PLR in that the taxpayer is drawing a bright line on a calendar in 1998 and says I want to plead guilty to the application of penalties on all prior distributions, essentially dis-posing of the retrospective portion of the issue. The taxpayer goes on to say that he really wants to “start a fresh” with a new program, essentially pretending that the old program did not exist by virtue of paying the penalties and interest on the pre-1998 withdrawals. Implicitly, by drawing this line on the calendar, the taxpayer does not want his new or “prospective” SEPP program to be shackled by or compared to the old program. Finally, the taxpayer describes his new program as being a variable one that recalculates each January with updated account balances and an updated life expectancy. Evidently the timing of money was more important to this taxpayer than absolute amounts in that the penalties for distributions in 1993 through 1997 approximate $100,000. On the other hand, this taxpayer’s annual distributions immediately jumped from $176,500 per year to $672,000 in 1998, plus the ability to withdraw more in 1999 and beyond on the presumption that the taxpayer’s successful in-vesting style continues.

In summary, it appears that recalculation is a close issue. Let’s revisit our original definition: “A series of payments or withdrawals from a deferred account that can either be exactly equal, or poten-tially somewhat dissimilar, that occur annually or more frequently and continue for a minimum time period that is the greater of five years or until you reach age 59½.” This definition no longer appears to be correct. The whole concept of recalculation, particularly when using updated account balances,124 seems to violate intellectual notion of what a substantially equal payment” should be.

122 See PLRs 97-05033, 98-18055, 98-21056, and 99-43050. 123 See PLR 99-09059. 124 And every positive PLR on this issue included the updating of new account balances.

“...Taxpayer A desires to modify the basis on which periodic distributions will continue to be made.… The proposed method... will be to calculate an annual payment by amortizing the account balance (Determined as of the third Monday in each January...[using a] life expectancy...[from]...Table V, §1.72-9 of the regulations...and reducing that figure by one for each subsequent year.… A constant annual interest rate of 7.2% will be assumed.… Taxpayer A realizes that this action will amount to a modification...of the withdrawals commenced in 1993, and will require the payment of the additional 10% penalty, plus interest for the...[prior year]...withdrawals...pursuant to §72(t)(4).”

Private Letter Ruling 99-09059

52

However, we can potentially solve this dilemma by adding one word, “system.” Thus, we could now say “SEPPSs” or “substantially equal periodic payment system.” The noun is now “system” as opposed to word “payment,” which is now an adjective. As a result, the concepts of consistency apply to the system itself, as opposed to the resulting payment amount, which may be wildly different or in-consistent depending on the variable values. If we return to the 12 PLRs and phrase our thinking or question a little differently, the results become much clearer and consistent:

• How many times did a taxpayer propose a SEPPS to be applied prospectively on a consistent basis and was ruled favorably by the IRS? The answer is 8 - 0.

• How many times did a taxpayer propose a SEPPS to be applied retrospectively and prospec-tively on a consistent basis and was ruled favorably by the IRS? The answer is 0 - 4.

It appears that there are two central issues here—first, the IRS does not differentiate between a SEPP and a SEPPS; second, retroactive application is not allowed. Whether the IRS is right125 on these issues is not really relevant; it has spoken in this case on multiple occasions in a reasonably consistent manner. The two crucial points with this issue appear to be the use of prospective-only action (no retroactiv-ity) and the selective choice of positive language such that program consistency is sold and the fact that the system to be applied contains variability almost becomes an incidental comment. In conclu-sion, it is author’s opinion that recalculation in conjunction with either the amortization or annuity method is still an open issue and should therefore be implemented carefully.

Fortunately, the IRS, in addition to the PLRs discussed, has directly addressed this issue in an in-formation letter:126

125 On a theoretical level, this author’s opinion is that the IRS has incorrectly liberalized the concept of a SEPP to a SEPPS; rather, that the original congressional intent was for the actual payments to be substantially equal, not that the system, which determines the payments, be substantially equal. That being said, the author is certainly willing to live with the out-come in that it adds a very effective tool to the taxpayer’s ability to build SEPPs that meet the taxpayer’s needs. 126 As defined in Revenue Procedure 2000-4, an information letter “calls attention to a well established interpretation or principle of tax law...without applying it to a specific set of facts.… An information letter is advisory only and has no bind-ing effect on the Service.” As a result, we can rely on the contents of an information letter from a theory perspective only and must carefully apply that theory to each individual’s unique fact set.

“A second method described in Q&A-12 [of Notice 89-25] determines an annual distribution amount by amortizing the taxpayer’s account balance over a number of years equal to the life expectancy of the account owner or the joint life and last survivor expectancy of the account owner and beneficiary at an interest rate that does not exceed a reasonable in-terest rate on the date payments commence. The resulting payment is the amount to be distributed each year. Again, the life expectancies to be used are found in Table V (one life) or Table VI (two lives) of Section 1.72-9 of the regulations. If payments are recalculated each year using the amortization method, then payments would be calculated in the same manner, using the account balance as of the same day of the year, the applicable life expectancy (or life expec-tancies), and the same interest rate ‘standard’ in effect for the same period of the year, which must provide an interest rate that does not exceed a reasonable interest rate on the date payments commence.” The above is an excerpt from an information letter received by the author from the IRS; more specifically the manager of the Employee Plans Actuarial Group 2, Tax Exempt and Government Entities Division of the Assistant General Coun-sel’s Office. This document goes on to discuss the annuity method with identical “recalculation” language, but for the obvious computational differences between the methods.

53

This document is an informal but explicit blessing of annual recalculation for use in conjunction with either the amortization or annuity methods, provided one carefully reads and carefully follows IRS advice. What does it really say?

• All three variables must be updated127 when recalculating.

• All three variables must be updated as of the same day each year.

• No methodology changes are permitted, simply a substitution of new values and the resultant computation.

• The interest rate standard that is updated annually must remain reasonable when the payments commence.

Within the context of the last point above, the IRS is talking about “within the year” not when the SEPP payment stream may have really commenced some years previously. Thus, you might compute your recalculation as of 12/31/1999 using the long-term applicable federal rate in force for January of 2000, but you might not actually take a distribution until 8/2000, some eight months later. Is the rate you used in January still reasonable in August? Maybe, maybe not. As a result, this sentence fragment effectively creates a “short-term time trap” in which waiting to commence the payments for any sub-stantial time after the recalculation date increases the exposure of interest rate movement that would render the rate used as unreasonable. Fortunately, there is an easy solution—simply commence the payment stream (or single payment for the year) in close proximity, preferably within the same month as the recalculation date.

Finally, annual recalculation can get complex, particularly when multiple accounts and lives are being updated. To that end, this author suggests that anyone planning on a SEPP program including annual recalculation should write to himself a detailed letter. Amazingly, we all tend to forget details of what happened a year ago. In this situation, success is in the details. Here is some model language that I suggest that each taxpayer use: I, [your name], have elected to commence substantially equal periodic payments pursuant

to IRC §72(t)(2)(A)(iv). I am including specific IRAs, [list of accounts, numbers, and balances], to-taling [the aggregate amount] as of [the most recent month-end preceding the commencement of the SEPPs]. I am electing to use the [amortization or annuity] method as described in Notice 89-25. Using my attained age of [your highest specific attained age as of commencement tax year] and an interest rate assumption of [describe method of interest rate determination] and use of the [Table V, Table VI, or UP-1984 mortality table] results in an annual distribution for the calendar year 20XX of [inset actual dollar amount]. I intend to continue these substantially equal periodic payments commencing in [insert commencement month and year] and running uninterrupted to no less than [insert the month and year or actual date upon which you will attain the age of 59½ or the expiration of 5 years] so as to avoid the reapplication of the 10% early withdrawal penalty as required under IRC §72(t)(4)(A). I am further adopting the [amortization or annuity] method requiring that I [do or do not] recompute the substantially equal periodic payment dollar amount on each anniversary subsequent to [insert commencement month and year].

127 This is contrary to some older 1980’s private letter rulings wherein taxpayers have received successful rulings to annu-ally recalculate while holding one or two of the variables constant. Nonetheless, history is history and is not the current IRS position on this matter. Something about “tugging on Superman’s cape” comes to mind here.

54

As a result, on each and every annual anniversary of [insert commencement month and year], I will compute the required substantially equal periodic payment for that year by measuring my account balances as of [insert the date of the preceding month-end], adopting an updated in-terest rate assumption based on [specifically identify the interest rate and month to be used, e.g. the long-term applicable federal rate in force for the month of June], and using my then attained age in conjunction with the [Table V, Table VI or UP-1984] mortality table. This amount shall be withdrawn from one or more of my included accounts within 30 days of this annual computation. I further understand that noncompliance or deviation from the rules I have identified above will con-stitute a modification of the method resulting in reapplication of the 10% penalty and intervening statutory interest as prescribed by IRC §72(t)(4)(A).

At this point we have probably beaten recalculation on SEPPs to death. What are the advantages and disadvantages of implementing recalculation?

Advantages—First, and potentially foremost, the taxpayer can discard the entire earlier discussion on account exhaustion. Assuming that a SEPP program including recalculation will include the use of updated account balances, there is simply no way the account can ever be reduced to zero. Also, the taxpayer will enjoy bigger withdrawals earlier based on continued positive investment experience.

Disadvantages—However, the amount of the subsequent year distributions is both an advantage and disadvantage. It is a disadvantage in that poor investment results can drive the withdrawal amount into the basement, almost zero but never actually zero. This downside variability will clearly not sit well with the taxpayer’s personal cash flow; it is simply a matter of degree.

In summary, recalculation removes account exhaustion and replaces it with volatility. Each tax-payer will need to decide individually if this trade is worth it.

Death and Divorce Unfortunately both death and divorce are a way of life. How does the IRS treat these issues if SEPPs are already in progress? If the taxpayer receiving the SEPPs dies, then per statute, §72(t)(2)(A)(ii), the SEPPs cease and no penalties are assessed.

If the taxpayer’s spouse dies, the IRS has ruled128 that the taxpayer can then recalculate using new methods and life expectancies. Thus, some flexibility is granted; however, the SEPPs must continue in some fashion.

If the taxpayer and his or her spouse divorce, the SEPPs stay with the taxpayer who was originally receiving the SEPPs. But, what if the assets within the SEPP universe are split (let’s say 50% and 50%) pursuant to a QDRO?129 The IRS has ruled130 that the SEPPs can continue with each spouse continuing with 50% of the original SEPP amount from the now-separated accounts. This is essentially a “step-in-stride” ruling. Further, the IRS has ruled that when the corpus of the IRA is split, the original owner of the IRA performing SEPPs may ratably reduce his or her SEPPs going forward131. Note that the IRA 128 See PLRs 89-19052 and 90-47076. 129 A “QDRO” is a qualified domestic relations order. 130 See PLR 97-39044. 131 See PLRs 00-50046, 00-52039, and 01-16056.

55

owner is not afforded an opportunity to switch methods, but rather simply a ratable reduction equal that portion of the IRA that was transferred to the ex-spouse. As an example, our taxpayer has an IRA with $XXX,XXX in it as of some date and is on a SEPP program of $60,000 per year in fixed with-drawals. As a result of a QDRO, our taxpayer is forced to transfer 40% of the IRA assets into a new IRA in his ex-spouse’s name leaving him 60% of the principal. In this case our taxpayer can ratably reduce his remaining withdrawals from $60,000 down to $36,000 per year. Further, the new owner/ex-spouse of the newly created IRA has three options:

• One, voluntarily pick up the remainderman $24,000 of SEPPs per year. In this case, the history of prior-year SEPPs would accrue to the new IRA owner; e.g. suppose the $60,000 per year had already been running for three years; then the new IRA owner would “so to speak” receive three years of credit toward satisfying the five year rule. However, the new IRA owner’s age would prevail in making the age 59½ test.

• Two, the new IRA owner may choose to do nothing at all.

• Three, the new IRA owner may choose to initiate a new SEPP stream completely independent of the former.

All this being said, these are just three newly released private letter rulings. As a result, a taxpayer facing these or similar circumstances would be well advised to seek professional help potentially re-sulting in a new PLR submission.

Simultaneous Processing of Other Exemptions §72(t) has been amended several times to the point where it now contains 12 different exceptions ranging from the originals—death and disability—to a new group of exceptions for new home pur-chase, excessive medical expense, and educational costs. How does the application of §72(t)(2)(A)(iv) sit with the simultaneous application of another exception on the same accounts? The concept of a frozen account universe for SEPPs and the only transaction permitted being the SEPP withdrawal would seem to argue against the processing of another exception transaction against the SEPP accounts. On the other hand, there is no statutory provision prohibiting such an action. Therefore, why shouldn’t a taxpayer be able to avail himself of such a transaction? As usual, the answer is unclear. Further, the IRS has never ruled on this particular issue, leaving us in virgin territory. As a result, this author would suggest the path of conservatism. Either don’t attempting such a transaction, or pursue a separate private letter ruling on the subject before executing such a transaction.

Planning of the SEPP The reader now has all of the strategic and tactical implementation tools132 necessary to effectively implement a SEPP or series of SEPPs to meet a taxpayer’s cash flow needs.

This author would like to suggest that the reader think about cash flow needs like a layer-caked matrix with the Y axis representing dollars and the X axis representing time. Returning to John for a

132 The author is the first person to admit that detailed discussions on SEPPs are pretty dry. As a result, to the reader who has made it this far, congratulations. You now know more about this subject that 90% of all professionals in the taxation arena.

56

moment, recall that he needed a cash flow increase commencing in the fourth year to cover anticipated college education expenses. Thus, one potential solution for John is:

• SEPP #1 commences immediately using either the amortization or annuity methods using no recalculation and assuming a 3% per year COLA. Assuming an interest rate of 8%, John needs to allocate approximately $700,000 (conveniently the sum of John’s IRA #1 and IRA #2) to this SEPP universe. Further, he can choose to leave IRAs in separate accounts, or he can com-bine them.

• SEPP #2 commences four years in the future using all or part of the remaining IRAs. Again, John would probably use the amortization or annuity method, but John has the luxury of post-poning all of his decision making until right before the commencement of program #2.

Tax Planning Issues for the Wealthy Are you wealthy? You might not think so; however, if you are reading this text, you have a high likelihood of fitting Congress’s definition of wealth. By one definition, Congress tells us that the wealthy commence at $65,500 of taxable income (for single individuals) or $109,250 of taxable in-come (for married couples) for tax year 2001. This is taxable income, not adjusted gross income, so we could probably set the limits for your gross income at $75,000 and $125,000 fairly easily. These are the income limits at which the 30.5% tax bracket kicks in. Interestingly, wealth in this case is an in-come test, not an assets test. Conventional financial wisdom says that IRAs, 401(k)s and the like are retirement vehicles that are typically off limits until age 60 or so. Making an early withdrawal to buy the Ferrari (or two trips around the world) is nothing but robbing from your own future. These are generalities that are most often true, but not always true. Let’s take an example. John, age 52, is still working, and plans to con-tinue working at a salary of $100,000 per year. He is married with two kids, two cars, a mortgage, and a front lawn to mow. As a result, his taxable income is $70,000 per year (about mid-bracket in the 27.5% tax bracket). Further, John already has $2,000,000 in deferred assets. Further, John can rea-sonably expect his $2 million to double in the next eight years when he wants to retire at age 60. Just to pick a number out of the air, John intends to start withdrawing 5% of his deferred assets or $200,000 per year starting eight years in the future. This will catapult John into the 35.5% tax bracket and poten-tially into the 39.1% tax bracket depending on his real taxable income. What should John do?

Although it is a bit counter-intuitive, John should start SEPPs now in the neighborhood of $30,000 per year. In no way is it being suggested that John spend the $30,000 of SEPPs; simply convert $30,000 of cash per year from a deferred-asset account to an after-tax investment account and imple-ment a long-term conservative investment strategy. Why? During the intervening eight years, John will pay 27.5% tax (or less) on the $30,000 per year or $8,250 per year in federal tax, leaving $21,750 per year to put in his after-tax investment account. If John waits until age 60, those same $30,000 in dol-lars will be taxed at 35.5% resulting in $10,650, which is $2,400 more in federal tax on the same dol-lars.

In short, this is an exercise in not leaving unused lower marginal tax brackets on the table in prior years. Unfortunately, John cannot wait to figure this out eight years in the future and then backtrack. John needs to be looking at the issue now, using some personal financial modeling and some intuition about the future to arrive at a likely financial and tax bracket outcome eight years in the future and fit a SEPP program into the picture, but only if the picture warrants it.

57

Conclusions Planning a SEPP program is not as easy as it might appear at first glance. However, the taxpayer should take the time to plan, think, and re-plan for as many potential scenarios as it takes to become comfortable with the outcomes. This iterative cycle, coupled with reasonable precautions will result in a SEPP program or series of SEPPs that will meet the taxpayer’s needs as well as escape the applica-tion of the 10% surtax and interest.

58

________________________________________________________________________________________________ Chapter 8 Risk Analysis Risk Assessment A taxpayer is exposed to two types of risk with SEPPs: 1. An individual’s SEPP program may be found to not comply with §72(t)(2)(A)(iv) at some point,

resulting in the application of §72(t)(4), e.g. the 10% surtax and intervening periodic interest. 2. If one’s SEPP program is challenged, the taxpayer may choose to expend the economic resources

to prevail against the challenge rather than pay the penalties.133 Each taxpayer can rather easily compute the dollar value of this risk based on his individual facts and circumstances. As an example, we discussed penalty computations back in the common concepts chapter. Because SEPPs must run for a minimum of five years, the minimum penalty—should it oc-cur—would approximate 58% of the taxpayer’s annual SEPP amount. Needless to say, this percentage grows with the passing of each year such that the penalty will approximate 140% by the tenth year.

What should a taxpayer do to mitigate this risk? The author would like to suggest that the taxpayer has three potentially different paths he can pursue—risk assumption, purchase of insurance (or risk transfer), or risk removal. Each of these tactics comes into play at differing levels of dollar risk as well as SEPP program aggressiveness.

Risk assumption means that the taxpayer personally assumes the risk of penalty. The taxpayer has two tools at his disposal should he choose to assume the risk. One, the taxpayer can intentionally de-sign a SEPP program of lesser dollar value than he might otherwise and therefore decrease the dollar value of corresponding risk. Two, the taxpayer can intentionally design a less risky (meaning less ag-gressive) program that would appear to be more safe and palatable to the IRS upon examination. Un-fortunately, individual risk tolerance is a very personal matter such that there are no pat answers. The best we can offer is that the taxpayer compute the maximum penalty and ask himself, “If I had to pay this penalty, could my personal finances withstand the payment and how would I feel about it?”

Risk transfer is the same as purchasing insurance. Although insurance companies do not issue “SEPP penalty” policies, there is something that is almost as good. All competent practitioners in the tax arena, both tax accountants and tax lawyers, invariably carry “errors and omissions” insurance. As a result, if the taxpayer so chooses, he can have the program designed by or reviewed by a licensed tax accountant and/or tax attorney. This design, and more importantly review, by the chosen professional should include a written opinion letter from that professional that indicates that he or she believes that the program as designed is in compliance with all provisions of §72(t). In the advent that the SEPP program is found not to comply, the taxpayer has recourse against the professional for having commit-ted a professional error or omission. Further, taxpayers should not hesitate to discuss this very issue with any professionals they chose to employ as well as discuss policy limits and how claims are made. Depending on the complexity of program design, professional tax accountants and lawyers may charge as little as $500 to as much as $10,000 for design, review, and opinion of a SEPP program.

Risk removal is elimination of hazard itself. The only way to remove the hazard is to go the Inter-nal Revenue Service and preclude them from finding your SEPP program noncompliant. The answer is 133 This author makes no attempt to quantify the costs of litigating on a SEPP program that has been challenged by the IRS other than to suggest that the cost of litigation could easily equal or exceed the initial cost of the penalties and interest.

59

a private letter ruling. As the title implies, it is a letter from the IRS to you (and only you) indicating the IRS has reviewed your facts, circumstances, planning and actions, and—in their opinion—rule that your SEPP plan is in compliance.134 In the event of an audit, the taxpayer need only produce the pri-vate letter ruling, and all IRS tax examiners will immediately proceed to the next topic of examination on the taxpayer’s return. Obtaining a PLR is not as onerous as one might think. The cost of obtaining a PLR is twofold—payment of IRS filing fees135 and payment to a professional tax accountant or tax at-torney to prepare the actual filing.136 The professional costs should be similar as indicated above. De-pending on the aggressiveness, and therefore level of risk, of a taxpayer’s proposed plan, it is not un-common for the professional to indicate that the plan risk factors are beyond his or her tolerance and therefore a private letter ruling is called for.

With these three tools for risk mitigation, which of the three should the taxpayer use? As usual, there is no clear-cut answer particularly taking into account the varying levels of taxpayer risk toler-ance. However, following is a chart of program design risks so that the taxpayer can get some sense of risk regarding his or her tentative program design.

134 This is the ultimate, perpetual “Get Out Of Jail Free” card. 135 For calendar-year 2000, filing fees on this issue are either $600 or $5,000 for under and over $200,000 of gross income, respectively. See Rev. Proc. 2000-8, Appendix A, Page 62. 136 This author suggests that individual taxpayers not attempt to prepare their own filings. A PLR filing is a highly regulated and stylized document. The addition of professional experience will create a better filing document, which will have a higher probability of success.

60

SEPP Program Design Risk Assessment Low Medium High Minimum Method With annual recalculation X Without annual recalculation X Use of Tables V or VI X Use of an alternate table X Amortization Method Without annual recalculation X With annual recalculation (as described) X With annual recalculation (modified) X Use of Tables V or VI X Use of an alternate table X Rate < LT/AFR * 1.2 X Rate > LT/AFR * 1.2 X Annuity Method Without annual recalculation X With annual recalculation (as described) X With annual recalculation (modified) X Use of a §1.401(a)(9) Table X Use of Table S (90CM or 80CNMST) X Use of an alternate table X Rate < LT/AFR * 1.2 X Rate > LT/AFR * 1.2 X Spousal Death or Divorce X Use of Multiple SEPP programs General X Use of new method X Use of new rate X Account Valuation Dates Year-, quarter-, or month-end X Other X Stub Period Payments X Account Universe Attempting to add to X Attempting to remove from X Cost of Living Increases Fixed 1% to 3% X Fixed greater than 3% X Relative X Retroactive Actions of Any Kind X Plan Interruption or Plan Modifications X New Computational Methods X

61

Needless to say, the allocation of any one action or feature into the low-, medium-, or high-risk category is a subjective process and the opinion of the author. Further, we would like to say that there is some magical scoring system to advise the taxpayer which strategies should be used at varying risk levels. Unfortunately, no such magic exists. Instead, each taxpayer will need to assess a combination of factors simultaneously: personal risk tolerance, dollar magnitude, and program risk level. Finally, many taxpayers have successfully played tax roulette in the past. This author ardently ad-vocates against such a strategy and instead suggests that each taxpayer who implements SEPPs should presume that he or she will be audited sometime during the SEPP period.

Individual Issues Many individuals are thoroughly capable to designing and implementing their own SEPP pro-grams; an equal number are not. Each taxpayer should self-evaluate and determine his or her abilities to go it alone versus seeking professional assistance. Generally, those individuals who are mathemati-cally facile, are comfortable with this text and the process, and have selected to implement a low- to medium-risk program will be fine on their own. An absence of confidence in any of these areas is probably cause to seek some level of outside assistance, either in whole or in part.

Selecting a Professional Many individuals will choose to either consult with a professional to obtain an outside opinion or may potentially turn over the whole project to the professional. How does the average taxpayer select a competent professional?

The average taxpayer, having read through the text to this point, knows more about substantially equal periodic payment programs and §72(t) issues than 90% or more of the practicing tax accountants and tax lawyers available in your local community.

With this in mind, the taxpayer really has an easy job of selecting a competent professional for help. It goes without saying that a professional should be both licensed and insured as well as have the apparent signals of permanency and stability. After that, it is time for some test questions. Ask ac-countants and lawyers137 if they have ever issued an opinion letter to a client on §72(t) issues. Ask if they have successfully submitted any private letter ruling requests. Ask if they have an opinion on the risk level associated with annual recalculation in conjunction with either the amortization or annuity methods. Ask if they have any thoughts on whatever you believe to be the most sensitive or riskiest design component of your program.

The answers you receive will, without question, be broad and, in some cases, test your own sense of amazement138. In short, just through the posing of two, three, or four questions, you will easily be able to differentiate between competent and less than competent professionals. Remember, as the cli-ent, you want to hire not only a skilled professional, but also a skilled professional who has a high

137 It is this author’s opinion that individuals seeking outside professional assistance should look for either a tax accountant or tax attorney as being the professions within which you will be able to find the requisite skills. However, other possibili-ties include Certified Financial Planners and Enrolled Agents. 138 This is the author’s method of forewarning readers that the concept of “bull-pucky” has never been limited to sales and marketing professions.

62

level of experience dealing with your particular needs. Finally, the best-fit accountant or lawyer will welcome you and your level of knowledge on SEPPs. You are his or her best client.

63

_______________________________________________________________________________________________

Chapter 9 Administration Record Keeping If there were ever a time for good record keeping, this is it. Once you have designed your SEPP program, you need to document it. It needs detail documentation for your own purposes, for your ac-countant’s or lawyer’s purposes if you use one, and for the IRS agent should the door bell ring. The easiest way to document your decisions is to send yourself a letter139 that outlines all of your decisions including all details along with copies of account statements that were pertinent at the time, mathe-matical and computational worksheets, and anything else that you think is pertinent.

Trustee Communications These communications should always be in writing. Even if a phone conversation or e-mail seems to have handled your request, always write a follow-up letter to the trustee140 confirming your agree-ment on steps and actions that are to occur. Unfortunately, the errors committed by the trustee are your errors. Normally, trustee errors are correctable, but not 100% of the time. Further, why invite problems and difficulties that might be even more difficult to explain at a later date, when a simple letter will suffice. Tax Matters The source from which you are withdrawing SEPP dollars must issue you a Form 1099-R (See Appendix I) on or before January 31st for the year just ended. Other than the amounts being correct, the most important box on the 1099-R is “Box 7: Distribution Code.” It should have a “2” in it. Referring to the reverse side of Form 1099-R, we find the distribution code definitions. A “2” means “Early dis-tribution exception applies (under age 59½ ) (You need not file Form 5329).” Conversely, if box 7 is blank, you will be required to complete Form 5329141 (See Appendix J) and attach it to your federal income tax return. Some trustees/custodians are willing to put the “2” in box 7 via a simple written request from the taxpayer. Take care of this early in the year. Do not wait until you receive your 1099-R, see that the “2” is not there, and then get on the phone. Other trustees/custodians will require more information be-fore they will comply with your request (another good reason to have written yourself a detailed SEPP program letter).

139 You will be amazed when you open this letter a year later to realize how much you will have forgotten in the intervening time period. 140 In this case, we are using the word “trustee” generically to mean whichever responsible party you might be dealing with to effect SEPP transactions. This may be an actual trustee, a custodian, a record keeper, a plan administrator, or your bro-ker. 141 If you are required to complete and file IRS Form 5329, please do not over look it. In the absence of the form and the absence of the “2” in box 7, IRS computers have a nasty habit of automatically issuing deficiency notices for 10% plus in-terest of the amount withdrawn.

64

SEPP withdrawals are always reportable as unearned ordinary income on lines 15a and 15b of IRS Form 1040.142 If your SEPP withdrawal is 100% taxable, then the same amount goes on both line 15a and 15b. If you have basis in the account(s) from which SEPP withdrawals are being made, place the total amount on line 15a and place the taxable amount on line 15b.

Taxes will be due because SEPP withdrawals are always reportable as ordinary income. How much tax will be due is entirely driven by the taxpayer’s total tax situation, not just the SEPP withdrawals. There are two ways for taxpayers to meet their tax obligations in this regard. One, most trustees offer to withhold federal and state taxes from your distribution. The taxpayer has the opportunity to avail himself of this process or decline. Second, taxpayers can always make quarterly estimated tax pay-ments143 at the federal level and state level as required. If the taxpayer chooses to make quarterly esti-mated payments, we suggest that SEPPs be withdrawn on a ratable quarterly basis to match the quar-terly estimated payment dates. This causes the distributions and the tax payments to line up nicely with no need to separately explain uneven distributions and the effect they would have on equally uneven estimated tax payments.

State Taxation At the federal level, SEPPs are always treated as taxable ordinary income. This is not always the case at the state level. Some states144 offer exclusions, credits, or otherwise afford the taxpayer favor-able tax treatment of SEPP distributions. The author is not suggesting that anyone should physically move simply to take advantage of favorable tax treatment. Rather, each taxpayer should take a quick glance at his or her state income tax instruction booklet to see if your state offers any tax-favored treatment.

The End My ultimate congratulations on your making it through. It was a tough haul, but I hope you will think it worthwhile. Needless to say, we attempted to think through SEPPs from every conceivable an-gle in order to anticipate your every question. It was my intent to have answered your first 97 ques-tions, hoping you would have only three remaining. I invite your criticisms as well as questions and recommendations for the next edition. The author can be reached at w.stecker@att.net (a.k.a. TheBadger on TheMotleyFool @ www.fool.com).

142 As of the 2000 IRS Form 1040. In future years, the correct reporting lines may move around depending on form design. 143 Estimated tax payments are due 15 calendar days after the close of the illogical IRS definition of a quarter: 4/15, 6/15, 10/15, and 1/15. Let’s see everyone count on her fingers and explain why 6/15 isn’t 7/15. 144 Unfortunately this is an ever-changing landscape; otherwise we would have been happy to provide a table of states with more favorable treatment.

65

________________________________________________________________________________________________ Appendix A Internal Revenue Code §72(t)

72(t) 10-PERCENT ADDITIONAL TAX ON EARLY DISTRIBUTIONS FROM QUALIFIED RETIREMENT PLANS.--

72(t)(1) IMPOSITION OF ADDITIONAL TAX.--If any taxpayer receives any amount from a qualified retirement plan (as defined in section 4974(c) ), the taxpayer’s tax under this chapter for the taxable year in which such amount is received shall be increased by an amount equal to 10 percent of the portion of such amount which is includible in gross income.

72(t)(2) SUBSECTION NOT TO APPLY TO CERTAIN DISTRIBUTIONS.--Except as provided in paragraphs (3) and (4), para-

graph (1) shall not apply to any of the following distributions:

72(t)(2)(A) IN GENERAL.--Distributions which are--

72(t)(2)(A)(i) made on or after the date on which the employee attains age 591/2, 72(t)(2)(A)(ii) made to a beneficiary (or to the estate of the employee) on or after the death of the employee, 72(t)(2)(A)(iii) attributable to the employee’s being disabled within the meaning of subsection (m)(7), 72(t)(2)(A)(iv) part of a series of substantially equal periodic payments (not less frequently than annually)

made for the life (or life expectancy) of the employee or the joint lives (or joint life expectancies) of such em-ployee and his designated beneficiary,

72(t)(2)(A)(v) made to an employee after separation from service after attainment of age 55, 72(t)(2)(A)(vi) dividends paid with respect to stock of a corporation which are described in section 404(k) , or 72(t)(2)(A)(vii) made on account of a levy under section 6331 on the qualified retirement plan.

72(t)(2)(B) MEDICAL EXPENSES.--Distributions made to the employee (other than distributions described in sub-paragraph (A), (C) or (D)) to the extent such distributions do not exceed the amount allowable as a deduction under section 213 to the employee for amounts paid during the taxable year for medical care (determined without regard to whether the employee itemizes deductions for such taxable year).

72(t)(2)(C) PAYMENTS TO ALTERNATE PAYEES PURSUANT TO QUALIFIED DOMESTIC RELATIONS ORDERS.--Any dis-

tribution to an alternate payee pursuant to a qualified domestic relations order (within the meaning of section 414(p)(1) ).

72(t)(2)(D) DISTRIBUTIONS TO UNEMPLOYED INDIVIDUALS FOR HEALTH INSURANCE PREMIUMS.--

72(t)(2)(D)(i) IN GENERAL.--Distributions from an individual retirement plan to an individual after separation from employment--

72(t)(2)(D)(i)(I) if such individual has received unemployment compensation for 12 consecutive weeks un-

der any Federal or State unemployment compensation law by reason of such separation, 72(t)(2)(D)(i)(II) if such distributions are made during any taxable year during which such unemployment

compensation is paid or the succeeding taxable year, and 72(t)(2)(D)(i)(III) to the extent such distributions do not exceed the amount paid during the taxable year for

insurance described in section 213(d)(1)(D) with respect to the individual and the individual’s spouse and de-pendents (as defined in section 152 ).

66

72(t)(2)(D)(ii) DISTRIBUTIONS AFTER REEMPLOYMENT.--Clause (i) shall not apply to any distribution made after the individual has been employed for at least 60 days after the separation from employment to which clause (i) ap-plies.

72(t)(2)(D)(iii) SELF-EMPLOYED INDIVIDUALS.--To the extent provided in regulations, a self-employed individ-

ual shall be treated as meeting the requirements of clause (i)(I) if, under Federal or State law, the individual would have received unemployment compensation but for the fact the individual was self-employed.

72(t)(2)(E) DISTRIBUTIONS FROM INDIVIDUAL RETIREMENT PLANS FOR HIGHER EDUCATION EXPENSES.--

Distributions to an individual from an individual retirement plan to the extent such distributions do not exceed the qualified higher education expenses (as defined in paragraph (7)) of the taxpayer for the taxable year. Distributions shall not be taken into account under the preceding sentence if such distributions are described in subparagraph (A), (C), or (D) or to the extent paragraph (1) does not apply to such distributions by reason of subparagraph (B).

72(t)(2)(F) DISTRIBUTIONS FROM CERTAIN PLANS FOR FIRST HOME PURCHASES.--Distributions to an individual

from an individual retirement plan which are qualified first-time homebuyer distributions (as defined in paragraph (8)). Distributions shall not be taken into account under the preceding sentence if such distributions are described in subparagraph (A), (C), (D), or (E) or to the extent paragraph (1) does not apply to such distributions by reason of subparagraph (B).

72(t)(3) LIMITATIONS.--

72(t)(3)(A) CERTAIN EXCEPTIONS NOT TO APPLY TO INDIVIDUAL RETIREMENT PLANS.--Subparagraphs (A)(v) and (C) of paragraph (2) shall not apply to distributions from an individual retirement plan.

72(t)(3)(B) PERIODIC PAYMENTS UNDER QUALIFIED PLANS MUST BEGIN AFTER SEPARATION.--Paragraph (2)(A)(iv)

shall not apply to any amount paid from a trust described in section 401(a) which is exempt from tax under section 501(a) or from a contract described in section 72(e)(5)(D)(ii) unless the series of payments begins after the employee separates from service.

72(t)(4) CHANGE IN SUBSTANTIALLY EQUAL PAYMENTS.--

72(t)(4)(A) IN GENERAL.--If--

72(t)(4)(A)(i) paragraph (1) does not apply to a distribution by reason of paragraph (2)(A)(iv), and 72(t)(4)(A)(ii) the series of payments under such paragraph are subsequently modified (other than by reason of

death or disability)--

72(t)(4)(A)(ii)(I) before the close of the 5-year period beginning with the date of the first payment and after the employee attains age 591/2, or

72(t)(4)(A)(ii)(II) before the employee attains age 591/2,

the taxpayer’s tax for the 1st taxable year in which such modification occurs shall be increased by an amount, deter-mined under regulations, equal to the tax which (but for paragraph (2)(A)(iv)) would have been imposed, plus interest for the deferral period.

72(t)(4)(B) DEFERRAL PERIOD.--For purposes of this paragraph, the term “deferral period” means the period be-ginning with the taxable year in which (without regard to paragraph (2)(A)(iv)) the distribution would have been in-cludible in gross income and ending with the taxable year in which the modification described in subparagraph (A) occurs.

72(t)(5) EMPLOYEE.--For purposes of this subsection, the term “employee” includes any participant, and in the case

of an individual retirement plan, the individual for whose benefit such plan was established. 72(t)(6) SPECIAL RULES FOR SIMPLE RETIREMENT ACCOUNTS.--In the case of any amount received from a simple re-

tirement account (within the meaning of section 408(p) ) during the 2-year period beginning on the date such individual

67

first participated in any qualified salary reduction arrangement maintained by the individual’s employer under section 408(p)(2) , paragraph (1) shall be applied by substituting “25 percent” for “10 percent”.

72(t)(7) QUALIFIED HIGHER EDUCATION EXPENSES.--For purposes of paragraph (2)(E)--

72(t)(7)(A) IN GENERAL.--The term “qualified higher education expenses” means qualified higher education ex-penses (as defined in section 529(e)(3) ) for education furnished to--

72(t)(7)(A)(i) the taxpayer, 72(t)(7)(A)(ii) the taxpayer’s spouse, or 72(t)(7)(A)(iii) any child (as defined in section 151(c)(3) ) or grandchild of the taxpayer or the taxpayer’s

spouse, at an eligible educational institution (as defined in section 529(e)(5) ).

72(t)(7)(B) COORDINATION WITH OTHER BENEFITS.--The amount of qualified higher education expenses for any taxable year shall be reduced as provided in section 25A(g)(2) .

72(t)(8) QUALIFIED FIRST-TIME HOMEBUYER DISTRIBUTIONS.--For purposes of paragraph (2)(F)--

72(t)(8)(A) IN GENERAL.--The term “qualified first-time homebuyer distribution” means any payment or distribu-tion received by an individual to the extent such payment or distribution is used by the individual before the close of the 120th day after the day on which such payment or distribution is received to pay qualified acquisition costs with respect to a principal residence of a first-time homebuyer who is such individual, the spouse of such individual, or any child, grandchild, or ancestor of such individual or the individual’s spouse.

72(t)(8)(B) LIFETIME DOLLAR LIMITATION.--The aggregate amount of payments or distributions received by an in-

dividual which may be treated as qualified first-time homebuyer distributions for any taxable year shall not exceed the excess (if any) of--

72(t)(8)(B)(i) $10,000, over 72(t)(8)(B)(ii) the aggregate amounts treated as qualified first-time homebuyer distributions with respect to

such individual for all prior taxable years.

72(t)(8)(C) QUALIFIED ACQUISITION COSTS.--For purposes of this paragraph, the term “qualified acquisition costs” means the costs of acquiring, constructing, or reconstructing a residence. Such term includes any usual or reasonable settlement, financing, or other closing costs.

72(t)(8)(D) FIRST-TIME HOMEBUYER; OTHER DEFINITIONS.--For purposes of this paragraph--

72(t)(8)(D)(i) FIRST-TIME HOMEBUYER.--The term “first-time homebuyer” means any individual if--

72(t)(8)(D)(i)(I) such individual (and if married, such individual’s spouse) had no present ownership interest in a principal residence during the 2-year period ending on the date of acquisition of the principal residence to which this paragraph applies, and

72(t)(8)(D)(i)(II) subsection (h) or (k) of section 1034 (as in effect on the day before the date of the enact-

ment of this paragraph) did not suspend the running of any period of time specified in section 1034 (as so in ef-fect) with respect to such individual on the day before the date the distribution is applied pursuant to subpara-graph (A).

72(t)(8)(D)(ii) PRINCIPAL RESIDENCE.--The term “principal residence” has the same meaning as when used in

section 121 . 72(t)(8)(D)(iii) DATE OF ACQUISITION.--The term “date of acquisition” means the date--

68

72(t)(8)(D)(iii)(I) on which a binding contract to acquire the principal residence to which subparagraph (A) applies is entered into, or

72(t)(8)(D)(iii)(II) on which construction or reconstruction of such a principal residence is commenced.

72(t)(8)(E) SPECIAL RULE WHERE DELAY IN ACQUISITION.--If any distribution from any individual retirement plan fails to meet the requirements of subparagraph (A) solely by reason of a delay or cancellation of the purchase or con-struction of the residence, the amount of the distribution may be contributed to an individual retirement plan as pro-vided in section 408(d)(3)(A)(i) (determined by substituting “120th day” for “60th day” in such section), except that--

72(t)(8)(E)(i) section 408(d)(3)(B) shall not be applied to such contribution, and 72(t)(8)(E)(ii) such amount shall not be taken into account in determining whether section 408(d)(3)(B) applies

to any other amount. [Caution: Code Sec. 72(t)(9), below, as added by P.L. 107-16, generally applies to distributions after December 31, 2001. For sunset provision, see P.L. 107-16, §901, in the amendment notes. ]

72(t)(9) SPECIAL RULE FOR ROLLOVERS TO SECTION 457 PLANS.--For purposes of this subsection, a distribution from

an eligible deferred compensation plan (as defined in section 457(b) ) of an eligible employer described in section 457(e)(1)(A) shall be treated as a distribution from a qualified retirement plan described in 4974(c)(1) to the extent that such distribution is attributable to an amount transferred to an eligible deferred compensation plan from a qualified re-tirement plan (as defined in section 4974(c) ).

69

________________________________________________________________________________________________ Appendix B

IRC Regulation §1.72-17A(f) Regulations On Determining Disability

Meaning of disabled.--(1) Section 72(m)(7) provides that an individual shall be considered to be disabled if he is unable to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or to be of long-continued and indefinite duration. In determining whether an individual’s impairment makes him unable to engage in any substantial gainful activity, primary consideration shall be given to the na-ture and severity of his impairment. Consideration shall also be given to other factors such as the individual’s education, training, and work experience. The substantial gainful activity to which section 72(m)(7) refers is the activity, or a compa-rable activity, in which the individual customarily engaged prior to the arising of the disability or prior to retirement if the individual was retired at the time the disability arose.

(2) Whether or not the impairment in a particular case constitutes a disability is to be determined with reference to all

the facts in the case. The following are examples of impairments which would ordinarily be considered as preventing sub-stantial gainful activity:

(i) Loss of use of two limbs; (ii) Certain progressive diseases which have resulted in the physical loss or atrophy of a limb, such as diabetes,

multiple sclerosis, or Buerger’s disease; (iii) Diseases of the heart, lungs, or blood vessels which have resulted in major loss of heart or lung reserve as evi-

denced by X-ray, electrocardiogram, or other objective findings, so that despite medical treatment breathlessness, pain, or fatigue is produced on slight exertion, such as walking several blocks, using public transportation, or doing small chores;

(iv) Cancer which is inoperable and progressive; (v) Damage to the brain or brain abnormality which has resulted in severe loss of judgment, intellect, orientation,

or memory; (vi) Mental diseases (e.g., psychosis or severe psychoneurosis) requiring continued institutionalization or constant

supervision of the individual; (vii) Loss or diminution of vision to the extent that the affected individual has a central visual acuity of no better

than 20/200 in the better eye after best correction, or has a limitation in the fields of vision such that the widest diameter of the visual fields subtends an angle no greater than 20 degrees;

(viii) Permanent and total loss of speech; (ix) Total deafness uncorrectible by a hearing aid.

The existence of one or more of the impairments described in this subparagraph (or of an impairment of greater severity) will not, however, in and of itself always permit a finding that an individual is disabled as defined in section 72(m)(7). Any impairment, whether of lesser or greater severity, must be evaluated in terms of whether it does in fact prevent the individ-ual from engaging in his customary or any comparable substantial gainful activity.

(3) In order to meet the requirements of section 72(m)(7), an impairment must be expected either to continue for a long and indefinite period or to result in death. Ordinarily, a terminal illness because of disease or injury would result in disability. The term “indefinite” is used in the sense that it cannot reasonably be anticipated that the impairment will, in the foresee-able future, be so diminished as no longer to prevent substantial gainful activity. For example, an individual who suffers a bone fracture which prevents him from working for an extended period of time will not be considered disabled, if his re-covery can be expected in the foreseeable future; if the fracture persistently fails to knit, the individual would ordinarily be considered disabled.

70

________________________________________________________________________________________________

Appendix C

Excerpt from Notice 89-25, 1989-1 C.B. 662 Q-12: In the case of an IRA or individual account plan, what constitutes a series of substantially equal periodic payments for purposes of section 72(t)(2)(A)(iv)? A-12: Section 72(t)(1) imposes an additional tax of 10 percent on the portion of early distributions from qualified retire-ment plans (including IRAs) includible in gross income. However, section 72(t)(2)(A)(iv) provides that this tax shall not apply to distributions which are part of a series of substantially equal periodic payments (not less frequently than annually) made for the life (or life expectancy) of the employee or the joint lives (or joint life expectancies) of the employee and beneficiary. Section 72(t)(4) provides that, if the series of periodic payments is subsequently modified within five years of the date of the first payment, or, if later, age 591/2, the exception to the 10 percent tax under section 72(t)(2)(A)(iv) does not apply, and the taxpayer's tax for the year of modification shall be increased by an amount, determined under regula-tions, which (but for the 72(t)(2)(A)(iv) exception) would have been imposed, plus interest. Payments will be considered to be substantially equal periodic payments within the meaning of section 72(t)(2)(A)(iv) if they are made according to one of the methods set forth below. Payments shall be treated as satisfying section 72(t)(2)(A)(iv) if the annual payment is determined using a method that would be acceptable for purposes of calculating the minimum distribution required under section 401(a)(9). For this pur-pose, the payment may be determined based on the life expectancy of the employee or the joint life and last survivor expec-tancy of the employee and beneficiary. Payments will also be treated as substantially equal periodic payments within the meaning of section 72(t)(2)(A)(iv) if the amount to be distributed annually is determined by amortizing the taxpayer's account balance over a number of years equal to the life expectancy of the account owner or the joint life and last survivor expectancy of the account owner and benefici-ary (with life expectancies determined in accordance with proposed section 1.401(a)(9)-1 of the regulations) at an interest rate that does not exceed a reasonable interest rate on the date payments commence. For example, a 50 year old individual with a life expectancy of 33.1, having an account balance of $100,000, and assuming an interest rate of 8 percent, could satisfy section 72(t)(2)(A)(iv) by distributing $8,679 annually, derived by amortizing $100,000 over 33.1 years at 8 percent interest. Finally, payments will be treated as substantially equal periodic payments if the amount to be distributed annually is deter-mined by dividing the taxpayer's account balance by an annuity factor (the present value of an annuity of $1 per year be-ginning at the taxpayer's age attained in the first distribution year and continuing for the life of the taxpayer) with such an-nuity factor derived using a reasonable mortality table and using an interest rate that does not exceed a reasonable interest rate on the date payments commence. If substantially equal monthly payments are being determined, the taxpayer's account balance would be divided by an annuity factor equal to the present value of an annuity of $1 per month beginning at the taxpayer's age attained in the first distribution year and continuing for the life of the taxpayer. For example, if the annuity factor for a $1 per year annuity for an individual who is 50 years old is 11.109 (assuming an interest rate of 8 percent and using the UP-1984 Mortality Table), an individual with a $100,000 account balance would receive an annual distribution of $9,002 ($100,000/11.109 = $9,002).

71

________________________________________________________________________________________________ Appendix D Excerpt from IRS Publication 590 (For Use In Preparing 1998 Returns) When Can I Withdraw or Use IRA Assets?

Because a traditional IRA is a tax-favored means of saving for your retirement, there are rules limiting the withdrawal and use of your IRA assets. Violation of the rules generally results in additional taxes in the year of violation. See What Acts Result in Penalties? later.

Note. These rules also apply to SIMPLE retirement accounts, which are discussed in chapter 5.

Age 59½ Rule

Generally, until you reach age 59½, you must pay a 10% additional tax if you withdraw assets (money or other property) from your traditional IRA. This tax is 10% of the part of the distribution that you have to include in gross income. It is in addition to any regular income tax on the amount you have to include in gross income. A number of exceptions to this rule are discussed below under Exceptions. Also see Premature Distributions (Early Withdrawals), later.

You may have to pay a 25% additional tax, rather than 10%, if you withdraw amounts from a SIMPLE IRA before you are age 59½ . See Additional Tax on Premature Distributions (Early Withdrawals), in chapter 5.

Note. If you receive a distribution from a traditional IRA that includes a return of nondeductible contributions, the 10% additional tax does not apply to the nontaxable part of the distribution. See Figuring the Nontaxable and Taxable Amounts under Are Distributions From My Traditional IRA Taxable?, later in this chapter.

After age 59½ and before age 70½. You can withdraw assets from your traditional IRA after you reach age 59½ without penalty. Even though you can, you do not have to withdraw any assets from your IRA until you reach age 70½. See When Must I Withdraw IRA Assets (Required Distributions)?, later in this chapter.

Exceptions

The exceptions to the age 59½ rule for distributions from traditional IRAs are in part designed to provide relief from hard-ship situations such as:

Significant unreimbursed medical expenses,

Paying medical insurance premiums after losing your job,

Disability, and

Death.

But there are also exceptions for distributions:

That are a part of a series of substantially equal payments (see Annuity, later),

That are not more than qualified higher education expenses (see Higher education expenses, later), and

To pay certain qualified first-time homebuyer amounts (see First home, later).

Note. Distributions that are rolled over, as discussed earlier, are not subject to regular income tax or the 10% addi-tional tax. Certain withdrawals of contributions are also tax-free and not subject to the 10% additional tax (see Tax-free withdrawal of contributions, later).

Unreimbursed medical expenses. Even if you are under age 59½, you do not have to pay the 10% tax on amounts you withdraw that are not more than:

72

The amount you paid for unreimbursed medical expenses during the year of the withdrawal, minus

7.5% of your adjusted gross income for the year of the withdrawal.

You can only take into account unreimbursed medical expenses that you would be able to include in figuring a de-duction for medical expenses on Schedule A, Form 1040. You do not have to itemize your deductions to take ad-vantage of this exception to the 10% additional tax.

Medical insurance. Even if you are under age 59½, you may not have to pay the 10% tax on amounts you with-draw from your traditional IRA during the year that are not more than the amount you paid during the year for medical insurance for yourself, your spouse, and your dependents. You will not have to pay the tax on these amounts if all four of the following conditions apply.

You lost your job.

You received unemployment compensation paid under any federal or state law for 12 consecutive weeks.

You make the withdrawals during either the year you received the unemployment compensation or the following year.

You make the withdrawals no later than 60 days after you have been reemployed.

Disability. If you become disabled before you reach age 59½, any amounts you withdraw from your traditional IRA because of your disability are not subject to the 10% additional tax.

You are considered disabled if you can furnish proof that you cannot do any substantial gainful activity because of your physical or mental condition. A physician must determine that your condition can be expected to result in death or to be of long continued and indefinite duration.

Death. If you die before reaching age 59½, the assets in your traditional IRA can be distributed to your beneficiary or to your estate without either having to pay the 10% additional tax.

However, if you inherit a traditional IRA from your deceased spouse and elect to treat it as your own (as discussed under Inherited IRAs earlier), any distribution you later receive before you reach age 59½ may be subject to the 10% additional tax.

Annuity. You can receive distributions from your traditional IRA that are part of a series of substantially equal payments over your life (or your life expectancy), or over the lives (or joint life expectancies) of you and your beneficiary, without having to pay the 10% additional tax, even if you receive such distributions before you are age 59½. You must use an IRS-approved distribution method and you must take at least one distribution annually for this exception to apply. See Figuring the Minimum Distribution, later, for one IRS-approved distribution method, generally referred to as the "life expectancy method." Unlike for minimum distribution purposes, this method, when used for this purpose, results in the exact amount required, not the minimum amount.

There are two other IRS-approved distribution methods that you can use. They are generally referred to as the "amortization method" and the "annuity factor method." These two methods are not discussed in this publication because they are more complex and generally require professional assistance. See IRS Notice 89-25 in Internal Revenue Cumulative Bulletin 1989-1 for more information on these two methods. This notice can be read in many libraries and IRS offices.

The payments under this exception must continue for at least 5 years, or until you reach age 59½, whichever is the longer period. This 5-year rule does not apply if a change from an approved distribution method is made because of the death or disability of the IRA owner.

If the payments under this exception are changed before the end of the above required periods for any reason other than the death or disability of the IRA owner, he or she will be subject to the 10% additional tax.

For example, if you received a lump-sum distribution of the balance in your traditional IRA before the end of the required period for your annuity distributions and you did not receive it because you were disabled, you would be subject to the 10% additional tax. The tax would apply to the lump-sum distribution and all previous distributions made under the exception rule.

73

Higher education expenses. Even if you are under age 59½, if you paid expenses for higher education during the year, part (or all) of any withdrawal may not be subject to the 10% tax on early withdrawals. The part not subject to the tax is generally the amount that is not more than the qualified higher education expenses (defined later) for the year for education furnished at an eligible educational institution (defined later). The education must be for you, your spouse, or the children or grandchildren of you or your spouse.

When determining the amount of the withdrawal that is not subject to the 10% tax, include qualified higher educa-tion expenses paid with any of the following funds.

An individual's earnings.

A loan.

A gift.

An inheritance given to either the student or the individual making the withdrawal.

Personal savings (including savings from a qualified state tuition program).

Do not include expenses paid with any of the following funds.

Tax-free distributions from an education IRA.

Tax-free scholarships, such as a Pell grant.

Tax-free employer-provided educational assistance.

Any tax-free payment (other than a gift, bequest, or devise) due to enrollment at an eligible educational institution.

Qualified higher education expenses. Qualified higher education expenses are tuition, fees, books, supplies, and equipment required for the enrollment or attendance of a student at an eligible educational institution. In addition, if the individual is at least a half-time student, room and board are qualified higher education expenses.

Eligible educational institution. This is any college, university, vocational school, or other postsecondary educa-tional institution eligible to participate in the student aid programs administered by the Department of Education. It includes virtually all accredited, public, nonprofit, and proprietary (privately owned profit-making) postsecondary institutions. The educational institution should be able to tell you if it is an eligible educational institution.

First home. To qualify for penalty-free withdrawal treatment as a first-time homebuyer distribution, a distribution must meet the following requirements.

It must be used to pay qualified acquisition costs (defined later) before the close of the 120th day after the day you received it.

It must be used to pay qualified acquisition costs for the main home of a first-time homebuyer (defined later) who is any of the following.

Yourself.

Your spouse.

Your or your spouse's child.

Your or your spouse's grandchild.

Your or your spouse's parent or other ancestor.

When added to all your prior qualified first-time homebuyer distributions, if any, the total distributions cannot be more than $10,000.

If both husband and wife are first-time homebuyers (defined later), they each can withdraw up to $10,000 penalty-free for a first home.

Qualified acquisition costs. Qualified acquisition costs include the following items.

74

Costs of buying, building, or rebuilding a home.

Any usual or reasonable settlement, financing, or other closing costs.

First-time homebuyer. A first-time homebuyer is, generally, any individual (and his or her spouse, if married) who had no present ownership interest in a main home during the 2-year period ending on the date the individual acquires the main home to which these rules apply.

Date of acquisition. The date of acquisition is the date that:

The first-time homebuyer enters into a binding contract to buy the main home to which these rules apply, or

The building or rebuilding of the main home to which these rules apply begins.

Tax-free withdrawal of contributions. If you made IRA contributions for 1998, you can withdraw them tax free by the due date of your return. If you have an extension of time to file your return, you can withdraw them tax free by the extended due date. You can do this if both the following apply.

You did not take a deduction for the contributions you withdraw.

You also withdraw any interest or other income earned on the contributions. You must include in income any earn-ings on the contributions you withdraw. Include the earnings in income for the year in which you made the with-drawn contributions.

Generally, except for any part of a withdrawal that is a return of nondeductible contributions (basis), any with-drawal of your contributions after the due date (or extended due date) of your return will be treated as a taxable distribution. Another exception is the return of an excess contribution as discussed later under What Acts Result in Penalties?, later.

Premature withdrawals tax. The 10% additional tax on withdrawals made before you reach age 59½ does not ap-ply to these tax-free withdrawals of your contributions. However, your early withdrawal of interest or other in-come must be reported on Form 5329 and, unless the withdrawal qualifies as an exception to the age 59½ rule, it will be subject to this tax. See Premature Distributions (Early Withdrawals) under What Acts Result in Penalties?, later.

Excess contributions tax. If any part of these contributions is an excess contribution for 1997, it is subject to a 6% excise tax. You will not have to pay the 6% tax if any 1997 excess contribution was withdrawn by April 15, 1998 (plus extensions), and if any 1998 excess contribution is withdrawn by April 15, 1999. See Excess Contributions under What Acts Result in Penalties?, later.

You may be able to treat a contribution made to one type of IRA as having been made to a different type of IRA. This is called recharacterizing the contribution. See Recharacterizations in chapter 2 for more information.

75

_______________________________________________________________________________________________ Appendix E Table I (Publication 590: Single Life Expectancy)

Age Divisor Age Divisor 35 47.3 73 13.9 36 46.4 74 13.2 37 45.4 75 12.5 38 44.4 76 11.9 39 43.5 77 11.2 40 42.5 78 10.6 41 41.5 79 10.0 42 40.6 80 9.5 43 39.6 81 8.9 44 38.7 82 8.4 45 37.7 83 7.9 46 36.8 84 7.4 47 35.9 85 6.9 48 34.9 86 6.5 49 34.0 87 6.1 50 33.1 88 5.7 51 32.2 89 5.3 52 31.3 90 5.0 53 30.4 91 4.7 54 29.5 92 4.4 55 28.6 93 4.1 56 27.7 94 3.9 57 26.8 95 3.7 58 25.9 96 3.4 59 25.0 97 3.2 60 24.2 98 3.0 61 23.3 99 2.8 62 22.5 100 2.7 63 21.6 101 2.5 64 20.8 102 2.3 65 20.0 103 2.1 66 19.2 104 1.9 67 18.4 105 1.8 68 17.6 106 1.6 69 16.8 107 1.4 70 16.0 108 1.3 71 15.3 109 1.1 72 14.6 110 1.0

76

_______________________________________________________________________________________________ Appendix F Table II (Publication 590: Joint Life and Last Survivor Expectancy) Ages 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

35 54.0 53.5 53.0 52.6 52.2 51.8 51.4 51.1 50.8 50.5 50.2 50.0 49.7 49.5 49.3 36 53.5 53.0 52.5 52.0 51.6 51.2 50.8 50.4 50.1 49.8 49.5 49.2 49.0 48.8 48.5 37 53.0 52.5 52.0 51.5 51.0 50.6 50.2 49.8 49.5 49.1 48.8 48.5 48.3 48.0 47.8 38 52.6 52.0 51.5 51.0 50.5 50.0 49.6 49.2 48.8 48.5 48.1 47.8 47.5 47.3 47.0 39 52.2 51.6 51.0 50.5 50.0 49.5 49.1 48.6 48.2 47.8 47.5 47.2 46.8 46.6 46.3 40 51.8 51.2 50.6 50.0 49.5 49.0 48.5 48.1 47.6 47.2 46.9 46.5 46.2 45.9 45.6 41 51.4 50.8 50.2 49.6 49.1 48.5 48.0 47.5 47.1 46.7 46.3 45.9 45.5 45.2 44.9 42 51.1 50.4 49.8 49.2 48.6 48.1 47.5 47.0 46.6 46.1 45.7 45.3 44.9 44.5 44.2 43 50.8 50.1 49.5 48.8 48.2 47.6 47.1 46.6 46.0 45.6 45.1 44.7 44.3 43.9 43.6 44 50.5 49.8 49.1 48.5 47.8 47.2 46.7 46.1 45.6 45.1 44.6 44.1 43.7 43.3 42.9 45 50.2 49.5 48.8 48.1 47.5 46.9 46.3 45.7 45.1 44.6 44.1 43.6 43.2 42.7 42.3 46 50.0 49.2 48.5 47.8 47.2 46.5 45.9 45.3 44.7 44.1 43.6 43.1 42.6 42.2 41.8 47 49.7 49.0 48.3 47.5 46.8 46.2 45.5 44.9 44.3 43.7 43.2 42.6 42.1 41.7 41.2 48 49.5 48.8 48.0 47.3 46.6 45.9 45.2 44.5 43.9 43.3 42.7 42.2 41.7 41.2 40.7 49 49.3 48.5 47.8 47.0 46.3 45.6 44.9 44.2 43.6 42.9 42.3 41.8 41.2 40.7 40.2 50 49.2 48.4 47.6 46.8 46.0 45.3 44.6 43.9 43.2 42.6 42.0 41.4 40.8 40.2 39.7 51 49.0 48.2 47.4 46.6 45.8 45.1 44.3 43.6 42.9 42.2 41.6 41.0 40.4 39.8 39.3 52 48.8 48.0 47.2 46.4 45.6 44.8 44.1 43.3 42.6 41.9 41.3 40.6 40.0 39.4 38.8 53 48.7 47.9 47.0 46.2 45.4 44.6 43.9 43.1 42.4 41.7 41.0 40.3 39.7 39.0 38.4 54 48.6 47.7 46.9 46.0 45.2 44.4 43.6 42.9 42.1 41.4 40.7 40.0 39.3 38.7 38.1 55 48.5 47.6 46.7 45.9 45.1 44.2 43.4 42.7 41.9 41.2 40.4 39.7 39.0 38.4 37.7 56 48.3 47.5 46.6 45.8 44.9 44.1 43.3 42.5 41.7 40.9 40.2 39.5 38.7 38.1 37.4 57 48.3 47.4 46.5 45.6 44.8 43.9 43.1 42.3 41.5 40.7 40.0 39.2 38.5 37.8 37.1 58 48.2 47.3 46.4 45.5 44.7 43.8 43.0 42.1 41.3 40.5 39.7 39.0 38.2 37.5 36.8 59 48.1 47.2 46.3 45.4 44.5 43.7 42.8 42.0 41.2 40.4 39.6 38.8 38.0 37.3 36.6 60 48.0 47.1 46.2 45.3 44.4 43.6 42.7 41.9 41.0 40.2 39.4 38.6 37.8 37.1 36.3 61 47.9 47.0 46.1 45.2 44.3 43.5 42.6 41.7 40.9 40.0 39.2 38.4 37.6 36.9 36.1 62 47.9 47.0 46.0 45.1 44.2 43.4 42.5 41.6 40.8 39.9 39.1 38.3 37.5 36.7 35.9 63 47.8 46.9 46.0 45.1 44.2 43.3 42.4 41.5 40.6 39.8 38.9 38.1 37.3 36.5 35.7 64 47.8 46.8 45.9 45.0 44.1 43.2 42.3 41.3 40.5 39.7 38.8 38.0 37.2 36.3 35.5 65 47.7 46.8 45.9 44.9 44.0 43.1 42.2 41.3 40.4 69.6 38.7 37.9 37.0 36.2 35.4 66 47.7 46.7 45.8 44.9 44.0 43.1 42.2 41.3 40.4 39.5 38.6 37.8 36.9 36.1 35.2 67 47.6 46.7 45.8 44.8 43.9 43.0 42.1 41.2 40.3 39.4 38.5 37.7 36.8 36.0 35.1 68 47.6 46.7 45.7 44.8 43.9 42.9 42.0 41.1 40.2 39.3 38.4 37.6 36.7 35.8 35.0 69 47.6 46.6 45.7 44.8 43.8 42.9 42.0 41.1 40.2 39.3 38.4 37.5 36.6 35.7 34.9 70 47.5 46.6 45.7 44.7 43.8 42.9 41.9 41.0 40.1 39.2 38.3 37.4 36.5 35.7 34.8 71 47.5 46.6 45.6 44.7 43.8 42.8 41.9 41.0 40.1 39.1 38.2 37.3 36.5 35.6 34.7 72 47.5 46.6 45.6 44.7 43.7 42.8 41.9 40.9 40.0 39.1 38.2 37.3 36.4 35.5 34.6 73 47.5 46.5 45.6 44.6 43.7 42.8 41.8 40.9 40.0 39.0 38.1 37.2 36.3 35.4 34.6 74 47.5 46.5 45.6 44.6 43.7 42.7 41.8 40.9 39.9 39.0 38.1 37.2 36.3 35.4 34.5 75 47.4 46.5 45.5 44.6 43.6 42.7 41.8 40.8 39.9 39.0 38.1 37.1 36.2 35.3 34.5 76 47.4 46.5 45.5 44.6 43.6 42.7 41.7 40.8 39.9 38.9 38.0 37.1 36.2 35.3 34.4 77 47.4 46.5 45.5 44.6 43.6 42.7 41.7 40.8 39.8 38.9 38.0 37.1 36.2 35.3 34.4 78 47.4 46.4 45.5 44.5 43.6 42.6 41.7 40.7 39.8 38.9 38.0 37.0 36.1 35.2 34.3 79 47.4 46.4 45.5 44.5 43.6 42.6 41.7 40.7 39.8 38.9 37.9 37.0 36.1 35.2 34.3 80 47.4 46.4 45.5 44.5 43.6 42.6 41.7 40.7 39.8 38.8 37.9 37.0 36.1 35.2 34.2 81 47.4 46.4 45.5 44.5 43.5 42.6 41.6 40.7 39.8 38.8 37.9 37.0 36.0 35.1 34.2 82 47.4 46.4 45.4 44.5 43.5 42.6 41.6 40.7 39.8 38.8 37.9 36.9 36.0 35.1 34.2 83 47.4 46.4 45.4 44.5 43.5 42.6 41.6 40.7 39.8 38.8 37.9 36.9 36.0 35.1 34.2 84 47.4 46.4 45.4 44.5 43.5 42.6 41.6 40.7 39.8 38.8 37.8 36.9 36.0 35.1 34.2 85 47.4 46.4 45.4 44.5 43.5 42.6 41.6 40.7 39.8 38.8 37.8 36.9 36.0 35.1 34.1 86 47.3 46.4 45.4 44.5 43.5 42.5 41.6 40.6 39.8 38.8 37.8 36.9 36.0 35.0 34.1 87 47.3 46.4 45.4 44.5 43.5 42.5 41.6 40.6 38.7 38.7 37.8 36.9 35.9 35.0 34.1 88 47.3 46.4 45.4 44.5 43.5 42.5 41.6 40.6 38.7 38.7 37.8 36.9 35.9 35.0 34.1 89 47.3 46.4 45.4 44.4 43.5 42.5 41.6 40.6 38.7 38.7 37.8 36.9 35.9 35.0 34.1 90 47.3 46.4 45.4 44.4 43.5 42.5 41.6 40.6 38.7 38.7 37.8 36.9 35.9 35.0 34.1 91 47.3 46.4 45.4 44.4 43.5 42.5 41.6 40.6 38.7 38.7 37.8 36.8 35.9 35.0 34.1 92 47.3 46.4 45.4 44.4 43.5 42.5 41.6 40.6 38.7 38.7 37.8 36.8 35.9 35.0 34.1

77

Joint Life and Last Survivor Expectancy (continued) Ages 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64

35 49.2 49.0 48.8 48.7 48.6 36 48.4 48.2 48.0 47.9 47.7 37 47.6 47.4 47.2 47.0 46.9 38 46.8 46.6 46.4 46.2 46.0 39 46.0 45.8 45.6 45.4 45.2 40 45.3 45.1 44.8 44.6 44.4 41 44.6 44.3 44.1 43.9 43.6 42 43.9 43.6 43.3 43.1 42.9 43 43.2 42.9 42.6 42.4 42.1 44 42.6 42.2 41.9 41.7 41.4 45 42.0 41.6 41.3 41.0 40.7 46 41.4 41.0 40.6 40.3 40.0 47 40.8 40.4 40.0 39.7 39.3 48 40.2 39.8 39.4 39.0 38.7 49 39.7 39.3 38.8 38.4 38.1 50 39.2 38.7 38.3 37.9 37.5 51 38.7 38.2 37.8 37.3 36.9 52 38.3 37.8 37.3 36.8 36.4 53 37.9 37.3 36.8 36.3 35.8 54 37.5 36.9 36.4 35.8 35.3 55 37.1 36.5 35.9 35.4 34.9 34.4 33.9 33.5 33.1 32.7 32.3 32.0 31.7 31.4 31.1 56 36.8 36.1 35.6 35.0 34.4 33.9 33.4 33.0 32.5 32.1 31.7 31.4 31.0 30.7 30.4 57 36.4 35.8 35.2 34.6 34.0 33.5 33.0 32.5 32.0 31.6 31.2 30.8 30.4 30.1 29.8 58 36.1 35.5 34.8 34.2 33.6 33.1 32.5 32.0 31.5 31.1 30.6 30.2 29.9 29.5 29.2 59 35.9 35.2 34.5 33.9 33.3 32.7 32.1 31.6 31.1 30.6 30.1 29.7 29.3 28.9 28.6 60 35.6 34.9 34.2 33.6 32.9 32.3 31.7 31.2 30.6 30.1 29.7 29.2 28.8 28.4 28.0 61 35.4 34.6 33.9 33.3 32.6 32.0 31.4 30.8 30.2 29.7 29.2 28.7 28.3 27.8 27.4 62 35.1 34.4 33.7 33.0 32.3 31.7 31.0 30.4 29.9 29.3 28.8 28.3 27.8 27.3 26.9 63 34.9 34.2 33.5 32.7 32.0 31.4 30.7 30.1 29.5 28.9 28.4 27.8 27.3 26.9 26.4 64 34.8 34.0 33.2 32.5 31.8 31.1 30.4 29.8 29.2 28.6 28.0 27.4 26.9 26.4 25.9 65 34.6 33.8 33.0 32.3 31.6 30.9 30.2 29.5 28.9 28.2 27.6 27.1 26.5 26.0 25.5 66 34.4 33.6 32.9 32.1 31.4 30.6 29.9 29.2 28.6 27.9 27.3 26.7 26.1 25.6 25.1 67 34.3 33.5 32.7 31.9 31.2 30.4 29.7 29.0 28.3 27.6 27.0 26.4 25.8 25.2 24.7 68 34.2 33.4 32.5 31.8 31.0 30.2 29.5 28.8 28.1 27.4 26.7 26.1 25.5 24.9 24.3 69 34.1 33.2 32.4 31.6 30.8 30.1 29.3 28.6 27.8 27.1 26.5 25.8 25.2 24.6 24.0 70 34.0 33.1 32.3 31.5 30.7 29.9 29.1 28.4 27.6 26.9 26.2 25.6 24.9 24.3 23.7 71 33.9 33.0 32.2 31.4 30.5 29.7 29.0 28.2 27.5 26.7 26.0 25.3 24.7 24.0 23.4 72 33.8 32.9 32.1 31.2 30.4 29.6 28.8 28.1 27.3 26.5 25.8 25.1 24.4 23.8 23.1 73 33.7 32.8 32.0 31.1 30.3 29.5 28.7 27.9 27.1 26.4 25.6 24.9 24.2 23.5 22.9 74 33.6 32.8 31.9 31.1 30.2 29.4 28.6 27.8 27.0 26.2 25.5 24.7 24.0 23.3 22.7 75 33.6 32.7 31.8 31.0 30.1 29.3 28.5 27.7 26.9 26.1 25.3 24.6 23.8 23.1 22.4 76 33.5 32.6 31.8 30.9 30.1 29.2 28.4 27.6 26.8 26.0 25.2 24.4 23.7 23.0 22.3 77 33.5 32.6 31.7 30.8 30.0 29.1 28.3 27.5 26.7 25.9 25.1 24.3 23.6 22.8 22.1 78 33.4 32.5 31.7 30.8 29.9 29.1 28.2 27.4 26.6 25.8 25.0 24.2 23.4 22.7 21.9 79 33.4 32.5 31.6 30.7 29.9 29.0 28.2 27.3 26.5 25.7 24.9 24.1 23.3 22.6 21.8 80 33.4 32.5 31.6 30.7 29.8 29.0 28.1 27.3 26.4 25.6 24.8 24.0 23.2 22.4 21.7 81 33.3 32.4 31.5 30.7 29.8 28.9 28.1 27.2 26.4 25.5 24.7 23.9 23.1 22.3 21.6 82 33.3 32.4 31.5 30.6 29.7 28.9 28.0 27.2 26.3 25.5 24.6 23.8 23.0 22.3 21.5 83 33.3 32.4 31.5 30.6 29.7 28.8 28.0 27.1 26.3 25.4 24.6 23.8 23.0 22.2 21.4 84 33.2 32.3 31.4 30.6 29.7 28.8 27.9 27.1 26.2 25.4 24.5 23.7 22.9 22.1 21.3 85 33.2 32.3 31.4 30.5 29.6 28.8 27.9 27.0 26.2 25.3 24.5 23.7 22.8 22.0 21.3 86 33.2 32.3 31.4 30.5 29.6 28.7 27.9 27.0 26.1 25.3 24.5 23.6 22.8 22.0 21.2 87 33.2 32.3 31.4 30.5 29.6 28.7 27.8 27.0 26.1 25.3 24.4 23.6 22.8 21.9 21.1 88 33.2 32.3 31.4 30.5 29.6 28.7 27.8 27.0 26.1 25.2 24.4 23.5 22.7 21.9 21.1 89 33.2 32.3 31.4 30.5 29.6 28.7 27.8 26.9 26.1 25.2 24.4 23.5 22.7 21.9 21.1 90 33.2 32.3 31.3 30.5 29.6 28.7 27.8 26.9 26.1 25.2 24.3 23.5 22.7 21.8 21.0 91 33.2 32.2 31.3 30.4 29.5 28.7 27.8 26.9 26.0 25.2 24.3 23.5 22.6 21.8 21.0 92 33.2 32.2 31.3 30.4 29.5 28.6 27.8 26.9 26.0 25.2 24.3 23.5 22.6 21.8 21.0

78

Joint Life and Last Survivor Expectancy (continued) Ages 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79

65 25.0 24.6 24.2 23.8 23.4 23.1 22.8 22.5 22.2 22.0 66 24.6 24.1 23.7 23.3 22.9 22.5 22.2 21.9 21.6 21.4 67 24.2 23.7 23.2 22.8 22.4 22.0 21.7 21.3 21.0 20.8 68 23.8 23.3 22.8 22.3 21.9 21.5 21.2 20.8 20.5 20.2 69 23.4 22.9 22.4 21.9 21.5 21.1 20.7 20.3 20.0 19.6 70 23.1 22.5 22.0 21.5 21.1 20.6 20.2 19.8 19.4 19.1 71 22.8 22.2 21.7 21.2 20.7 20.2 19.8 19.4 19.0 18.6 72 22.5 21.9 21.3 20.8 20.3 19.8 19.4 18.9 18.5 18.2 73 22.2 21.6 21.0 20.5 20.0 19.4 19.0 18.5 18.1 17.7 74 22.0 21.4 20.8 20.2 19.6 19.1 18.6 18.2 17.7 17.3 75 21.8 21.1 20.5 19.9 19.3 18.8 18.3 17.8 17.3 16.9 16.5 16.1 15.8 15.4 15.1 76 21.6 20.9 20.3 19.7 19.1 18.5 18.0 17.5 17.0 16.5 16.1 15.7 15.4 15.0 14.7 77 21.4 20.7 20.1 19.4 18.8 18.3 17.7 17.2 16.7 16.2 15.8 15.4 15.0 14.6 14.3 78 21.2 20.5 19.9 19.2 18.6 18.0 17.5 16.9 16.4 15.9 15.4 15.0 14.6 14.2 13.9 79 21.1 20.4 19.7 19.0 18.4 17.8 17.2 16.7 16.1 15.6 15.1 14.7 14.3 13.9 13.5 80 21.0 20.2 19.5 18.9 18.2 17.6 17.0 16.4 15.9 15.4 14.9 14.4 14.0 13.5 13.2 81 20.8 20.1 19.4 18.7 18.1 17.4 16.8 16.2 15.7 15.1 14.6 14.1 13.7 13.2 12.8 82 20.7 20.0 19.3 18.6 17.9 17.3 16.6 16.0 15.5 14.9 14.4 13.9 13.4 13.0 12.5 83 20.6 19.9 19.2 18.5 17.8 17.1 16.5 15.9 15.3 14.7 14.2 13.7 13.2 12.7 12.3 84 20.5 19.8 19.1 18.4 17.7 17.0 16.3 15.7 15.1 14.5 14.0 13.5 13.0 12.5 12.0 85 20.5 19.7 19.0 18.3 17.6 16.9 16.2 15.6 15.0 14.4 13.8 13.3 12.8 12.3 11.8 86 20.4 19.6 18.9 18.2 17.5 16.8 16.1 15.5 14.8 14.2 13.7 13.1 12.6 12.1 11.6 87 20.4 19.6 18.8 18.1 17.4 16.7 16.0 15.4 14.7 14.1 13.5 13.0 12.4 11.9 11.4 88 20.3 19.5 18.8 18.0 17.3 16.6 15.9 15.3 14.6 14.0 13.4 12.8 12.3 11.8 11.3 89 20.3 19.5 18.7 18.0 17.2 16.5 15.8 15.2 14.5 13.9 13.3 12.7 12.2 11.6 11.1 90 20.2 19.4 18.7 17.9 17.2 16.5 15.8 15.1 14.5 13.8 13.2 12.6 12.1 11.5 11.0 91 20.2 19.4 18.6 17.9 17.1 16.4 15.7 15.0 14.4 13.7 13.1 12.5 12.0 11.4 10.9 92 20.2 19.4 18.6 17.8 17.1 16.4 15.7 15.0 14.3 13.7 13.1 12.5 11.9 11.3 10.8 93 20.1 19.3 18.6 17.8 17.1 16.3 15.6 14.9 14.3 13.6 13.0 12.4 11.8 11.3 10.7 94 20.1 19.3 18.5 17.8 17.0 16.3 15.6 14.9 14.2 13.6 12.9 12.3 11.7 11.2 10.6 95 20.1 19.3 18.5 17.8 17.0 16.3 15.6 14.9 14.2 13.5 12.9 12.3 11.7 11.1 10.6 96 20.1 19.3 18.5 17.7 17.0 16.2 15.5 14.8 14.2 13.5 12.9 12.2 11.6 11.1 10.5 97 20.1 19.3 18.5 17.7 17.0 16.2 15.5 14.8 14.1 13.5 12.8 12.2 11.6 11.0 10.5 98 20.1 19.3 18.5 17.7 16.9 16.2 15.5 14.8 14.1 13.4 12.8 12.2 11.5 11.0 10.4 99 20.0 19.2 18.5 17.7 16.9 16.2 15.5 14.7 14.1 13.4 12.7 12.1 11.5 10.9 10.4

100 20.0 19.2 18.4 17.7 16.9 16.2 15.4 14.7 14.0 13.4 12.7 12.1 11.5 10.9 10.3 101 20.0 19.2 18.4 17.7 16.9 16.1 15.4 14.7 14.0 13.3 12.7 12.1 11.4 10.8 10.3 102 20.0 19.2 18.4 17.6 16.9 16.1 15.4 14.7 14.0 13.3 12.7 12.0 11.4 10.8 10.2 103 20.0 19.2 18.4 17.6 16.9 16.1 15.4 14.7 14.0 13.3 12.6 12.0 11.4 10.8 10.2 104 20.0 19.2 18.4 17.6 16.9 16.1 15.4 14.7 14.0 13.3 12.6 12.0 11.4 10.8 10.2 105 20.0 19.2 18.4 17.6 16.8 16.1 15.4 14.6 13.9 13.3 12.6 12.0 11.3 10.7 10.2 106 20.0 19.2 18.4 17.6 16.8 16.1 15.4 14.6 13.9 13.3 12.6 11.9 11.3 10.7 10.1 107 20.0 19.2 18.4 17.6 16.8 16.1 15.3 14.6 13.9 13.2 12.6 11.9 11.3 10.7 10.1 108 20.0 19.2 18.4 17.6 16.8 16.1 15.3 14.6 13.9 13.2 12.6 11.9 11.3 10.7 10.1 109 20.0 19.2 18.4 17.6 16.8 16.1 15.3 14.6 13.9 13.2 12.6 11.9 11.3 10.7 10.1 110 20.0 19.2 18.4 17.6 16.8 16.1 15.3 14.6 13.9 13.2 12.6 11.9 11.3 10.7 10.1 111 20.0 19.2 18.4 17.6 16.8 16.0 15.3 14.6 13.9 13.2 12.6 11.9 11.3 10.7 10.1 112 20.0 19.2 18.4 17.6 16.8 16.0 15.3 14.6 13.9 13.2 12.5 11.9 11.3 10.6 10.1 113 20.0 19.2 18.4 17.6 16.8 16.0 15.3 14.6 13.9 13.2 12.5 11.9 11.2 10.6 10.0 114 20.0 19.2 18.4 17.6 16.8 16.0 15.3 14.6 13.9 13.2 12.5 11.9 11.2 10.6 10.0 115 20.0 19.2 18.4 17.6 16.8 16.0 15.3 14.6 13.9 13.2 12.5 11.9 11.2 10.6 10.0

79

______________________________________________________________________________________________ Appendix G UP-1984 Mortality Table

Noninsured Pensioner Mortality: The Up-1984 Table William W. Fellers & Paul H. Jackson

The Wyatt Company, Washington, D.C.

% Deaths Life Age Per Age Population Deaths Expectancy 15 0.1453% 77177281 112139 59.30 16 0.1437% 77065142 110743 58.38 17 0.1414% 76954400 108814 57.47 18 0.1385% 76845586 106431 56.55 19 0.1351% 76739155 103675 55.62 20 0.1311% 76635481 100469 54.70 21 0.1267% 76535011 96970 53.77 22 0.1219% 76438042 93178 52.84 23 0.1167% 76344864 89094 51.90 24 0.1149% 76255769 87618 50.96 25 0.1129% 76168151 85994 50.02 26 0.1107% 76082157 84223 49.08 27 0.1083% 75997934 82306 48.13 28 0.1058% 75915629 80319 47.18 29 0.1083% 75835310 82130 46.23 30 0.1111% 75753180 84162 45.28 31 0.1141% 75669019 86338 44.33 32 0.1173% 75582680 88658 43.38 33 0.1208% 75494022 91197 42.43 34 0.1297% 75402825 97797 41.48 35 0.1398% 75305027 105276 40.53 36 0.1513% 75199751 113777 39.59 37 0.1643% 75085974 123366 38.65 38 0.1792% 74962608 134333 37.71 39 0.1948% 74828275 145765 36.78 40 0.2125% 74682509 158700 35.85 41 0.2327% 74523809 173417 34.92 42 0.2556% 74350392 190040 34.00 43 0.2818% 74160352 208984 33.09 44 0.3095% 73951368 228879 32.18 45 0.3410% 73722489 251394 31.28

80

% Deaths Life

Age Per Age Population Deaths Expectancy 46 0.3769% 73471095 276913 30.39 47 0.4180% 73194183 305952 29.50 48 0.4635% 72888231 337837 28.62 49 0.5103% 72550394 370225 27.75 50 0.5616% 72180169 405364 26.89 51 0.6196% 71774806 444717 26.04 52 0.6853% 71330089 488825 25.20 53 0.7543% 70841264 534356 24.37 54 0.8278% 70306908 582001 23.55 55 0.9033% 69724907 629825 22.74 56 0.9875% 69095082 682314 21.95 57 1.0814% 68412768 739816 21.16 58 1.1863% 67672953 802804 20.39 59 1.2952% 66870149 866102 19.63 60 1.4162% 66004046 934749 18.88 61 1.5509% 65069297 1009160 18.14 62 1.7010% 64060137 1089663 17.42 63 1.8685% 62970474 1176603 16.51 64 2.0517% 61793871 1267825 16.02 65 2.2562% 60526046 1365589 15.35 66 2.4847% 59160458 1469960 14.69 67 2.7232% 57690498 1571028 14.05 68 2.9634% 56119470 1663044 13.43 69 3.2073% 54456426 1746581 12.82 70 3.4743% 52709845 1831298 12.23 71 3.7667% 50878547 1916442 11.65 72 4.0871% 48962104 2001130 11.09 73 4.4504% 46960974 2089951 10.54 74 4.8504% 44871023 2176424 10.01 75 5.2913% 42694599 2259099 9.49 76 5.7775% 40435500 2336161 9.00 77 6.3142% 38099339 2405668 8.52 78 6.8628% 35693670 2449585 8.06 79 7.4648% 33244085 2481604 7.62 80 8.1256% 30762481 2499636 7.19 81 8.8518% 28262844 2501770 6.78 82 9.6218% 25761074 2478679 6.39 83 10.4310% 23282395 2428587 6.02 84 11.2816% 20853808 2352643 5.66

81

% Deaths Life

Age Per Age Population Deaths Expectancy 85 12.2079% 18501165 2258604 5.32 86 13.2174% 16242561 2146844 4.99 87 14.3179% 14095717 2018211 4.67 88 15.5147% 12077506 1873789 4.37 89 16.8208% 10203717 1716347 4.08 90 18.2461% 8487371 1548614 3.80 91 19.8030% 6938756 1374082 3.54 92 21.5035% 5564675 1196600 3.29 93 23.2983% 4368075 1017687 3.05 94 25.2545% 3350388 846124 2.83 95 27.3878% 2504264 685863 2.62 96 29.7152% 1818401 540342 2.42 97 32.2553% 1278060 412242 2.23 98 34.9505% 865818 302608 2.05 99 37.8865% 563210 213381 1.88 100 41.0875% 349829 143736 1.72 101 44.5768% 206093 91870 1.57 102 48.3830% 114223 55265 1.43 103 52.4301% 58959 30912 1.29 104 56.8365% 28047 15941 1.17 105 61.6382% 12106 7462 1.05 106 66.8696% 4644 3105 0.94 107 72.5745% 1539 1117 0.84 108 78.6495% 422 332 0.75 109 85.2659% 90 77 0.66 110 92.4666% 13 12 0.58 111 100.0000% 1 1 0.00

82

_______________________________________________________________________________________________ Appendix H Sample Annuitant Divisor Table (Based on UP-1984)

Age

Published UP1984 Factor

UP1984 Raw

Mortality (1)

Computed

Factor

Difference

Probability of Living After Age Attain.

NPV of $1

@ X%

Annuitant

Factor

Percent To

Original

15 0.001453 77177281 0.001453 0.0000 0.0000 0.0000 0.0000 77.18%16 0.001437 77065142 0.001437 0.0000 0.0000 0.0000 0.0000 77.07%17 0.001414 76954400 0.001414 0.0000 0.0000 0.0000 0.0000 76.95%18 0.001385 76845586 0.001385 0.0000 0.0000 0.0000 0.0000 76.85%19 0.001351 76739155 0.001351 0.0000 0.0000 0.0000 0.0000 76.74%20 0.001311 76635480 0.001311 0.0000 0.0000 0.0000 0.0000 76.64%21 0.001267 76535011 0.001267 0.0000 0.0000 0.0000 0.0000 76.54%22 0.001219 76438042 0.001219 0.0000 0.0000 0.0000 0.0000 76.44%23 0.001167 76344864 0.001167 0.0000 0.0000 0.0000 0.0000 76.34%24 0.001149 76255769 0.001149 0.0000 0.0000 0.0000 0.0000 76.26%25 0.001129 76168151 0.001129 0.0000 0.0000 0.0000 0.0000 76.17%26 0.001107 76082157 0.001107 0.0000 0.0000 0.0000 0.0000 76.08%27 0.001083 75997934 0.001083 0.0000 0.0000 0.0000 0.0000 76.00%28 0.001058 75915629 0.001058 0.0000 0.0000 0.0000 0.0000 75.92%29 0.001083 75835310 0.001083 0.0000 0.0000 0.0000 0.0000 75.84%30 0.001111 75753180 0.001111 0.0000 0.0000 0.0000 0.0000 75.75%31 0.001141 75669018 0.001141 0.0000 0.0000 0.0000 0.0000 75.67%32 0.001173 75582680 0.001173 0.0000 0.0000 0.0000 0.0000 75.58%33 0.001208 75494022 0.001208 0.0000 0.0000 0.0000 0.0000 75.49%34 0.001297 75402825 0.001297 0.0000 0.0000 0.0000 0.0000 75.40%35 0.001398 75305027 0.001398 0.0000 0.0000 0.0000 0.0000 75.31%36 0.001513 75199751 0.001513 0.0000 0.0000 0.0000 0.0000 75.20%37 0.001643 75085974 0.001643 0.0000 0.0000 0.0000 0.0000 75.09%38 0.001792 74962608 0.001792 0.0000 0.0000 0.0000 0.0000 74.96%39 0.001948 74828275 0.001948 0.0000 0.0000 0.0000 0.0000 74.83%40 0.002125 74682509 0.002125 0.0000 0.0000 0.0000 0.0000 74.68%41 0.002327 74523809 0.002327 0.0000 0.0000 0.0000 0.0000 74.52%42 0.002556 74350392 0.002556 0.0000 0.0000 0.0000 0.0000 74.35%43 0.002818 74160352 0.002818 0.0000 0.0000 0.0000 0.0000 74.16%44 0.003095 73951368 0.003095 0.0000 0.0000 0.0000 0.0000 73.95%45 0.00341 73722489 0.00341 0.0000 0.0000 0.0000 0.0000 73.72%46 0.003769 73471095 0.003769 0.0000 0.0000 0.0000 0.0000 73.47%47 0.00418 73194183 0.00418 0.0000 0.0000 0.0000 0.0000 73.19%48 0.004635 72888231 0.004635 0.0000 0.0000 0.0000 0.0000 72.89%49 0.005103 72550394 0.005103 0.0000 0.0000 0.0000 0.0000 72.55%50 0.005616 72180169 0.005616 0.0000 0.0000 0.0000 0.0000 72.18%

83

Age

Published UP1984 Factor

UP1984 Raw

Mortality (1)

Computed

Factor

Difference

Probability of Living After Age Attain.

NPV of $1

@ X%

Annuitant

Factor

Percent To

Original

51 0.006196 71774805 0.006196 0.0000 0.0000 0.0000 0.0000 71.77%52 0.006853 71330089 0.006853 0.0000 1.0000 1.0000 1.0000 71.33%53 0.007543 70841264 0.007543 0.0000 0.9931 0.9259 0.9196 70.84%54 0.008278 70306908 0.008278 0.0000 0.9857 0.8573 0.8450 70.31%55 0.009033 69724907 0.009033 0.0000 0.9775 0.7938 0.7760 69.72%56 0.009875 69095082 0.009875 0.0000 0.9687 0.7350 0.7120 69.10%57 0.010814 68412768 0.010814 0.0000 0.9591 0.6806 0.6527 68.41%58 0.011863 67672953 0.011863 0.0000 0.9487 0.6302 0.5979 67.67%59 0.012952 66870149 0.012952 0.0000 0.9375 0.5835 0.5470 66.87%60 0.014162 66004046 0.014162 0.0000 0.9253 0.5403 0.4999 66.00%61 0.015509 65069297 0.015509 0.0000 0.9122 0.5002 0.4563 65.07%62 0.01701 64060137 0.01701 0.0000 0.8981 0.4632 0.4160 64.06%63 0.018685 62970474 0.018685 0.0000 0.8828 0.4289 0.3786 62.97%64 0.020517 61793871 0.020517 0.0000 0.8663 0.3971 0.3440 61.79%65 0.022562 60526046 0.022562 0.0000 0.8485 0.3677 0.3120 60.53%66 0.024847 59160458 0.024847 0.0000 0.8294 0.3405 0.2824 59.16%67 0.027232 57690498 0.027232 0.0000 0.8088 0.3152 0.2550 57.69%68 0.029634 56119470 0.029634 0.0000 0.7868 0.2919 0.2296 56.12%69 0.032073 54456426 0.032073 0.0000 0.7634 0.2703 0.2063 54.46%70 0.034743 52709845 0.034743 0.0000 0.7390 0.2502 0.1849 52.71%71 0.037667 50878547 0.037667 0.0000 0.7133 0.2317 0.1653 50.88%72 0.040871 48962104 0.040871 0.0000 0.6864 0.2145 0.1473 48.96%73 0.044504 46960974 0.044504 0.0000 0.6584 0.1987 0.1308 46.96%74 0.048504 44871023 0.048504 0.0000 0.6291 0.1839 0.1157 44.87%75 0.052913 42694599 0.052913 0.0000 0.5986 0.1703 0.1019 42.69%76 0.057775 40435500 0.057775 0.0000 0.5669 0.1577 0.0894 40.44%77 0.063142 38099339 0.063142 0.0000 0.5341 0.1460 0.0780 38.10%78 0.068628 35693670 0.068628 0.0000 0.5004 0.1352 0.0677 35.69%79 0.074648 33244085 0.074648 0.0000 0.4661 0.1252 0.0583 33.24%80 0.081256 30762481 0.081256 0.0000 0.4313 0.1159 0.0500 30.76%81 0.088518 28262844 0.088518 0.0000 0.3962 0.1073 0.0425 28.26%82 0.096218 25761074 0.096218 0.0000 0.3612 0.0994 0.0359 25.76%83 0.10431 23282395 0.10431 0.0000 0.3264 0.0920 0.0300 23.28%84 0.112816 20853808 0.112816 0.0000 0.2924 0.0852 0.0249 20.85%85 0.122079 18501165 0.122079 0.0000 0.2594 0.0789 0.0205 18.50%86 0.132174 16242561 0.132174 0.0000 0.2277 0.0730 0.0166 16.24%87 0.143179 14095717 0.143179 0.0000 0.1976 0.0676 0.0134 14.10%88 0.155147 12077506 0.155147 0.0000 0.1693 0.0626 0.0106 12.08%89 0.168208 10203717 0.168208 0.0000 0.1430 0.0580 0.0083 10.20%90 0.182461 8487371 0.182461 0.0000 0.1190 0.0537 0.0064 8.49%

84

Age

Published UP1984 Factor

UP1984 Raw

Mortality (1)

Computed

Factor

Difference

Probability of Living After Age Attain.

NPV of $1

@ X%

Annuitant

Factor

Percent To

Original

91 0.19803 6938756 0.19803 0.0000 0.0973 0.0497 0.0048 6.94%92 0.215035 5564675 0.215035 0.0000 0.0780 0.0460 0.0036 5.56%93 0.232983 4368075 0.232983 0.0000 0.0612 0.0426 0.0026 4.37%94 0.252545 3350388 0.252545 0.0000 0.0470 0.0395 0.0019 3.35%95 0.273878 2504264 0.273878 0.0000 0.0351 0.0365 0.0013 2.50%96 0.297152 1818401 0.297152 0.0000 0.0255 0.0338 0.0009 1.82%97 0.322553 1278060 0.322553 0.0000 0.0179 0.0313 0.0006 1.28%98 0.349505 865818 0.349505 0.0000 0.0121 0.0290 0.0004 0.87%99 0.378865 563210 0.378866 0.0000 0.0079 0.0269 0.0002 0.56%

100 0.410875 349829 0.410875 0.0000 0.0049 0.0249 0.0001 0.35%101 0.445768 206093 0.44577 0.0000 0.0029 0.0230 0.0001 0.21%102 0.48383 114223 0.483825 0.0000 0.0016 0.0213 0.0000 0.11%103 0.524301 58959 0.524297 0.0000 0.0008 0.0197 0.0000 0.06%104 0.568365 28047 0.568367 0.0000 0.0004 0.0183 0.0000 0.03%105 0.616382 12106 0.616389 0.0000 0.0002 0.0169 0.0000 0.01%106 0.668696 4644 0.668605 0.0001 0.0001 0.0157 0.0000 0.00%107 0.725745 1539 0.725796 -0.0001 0.0000 0.0145 0.0000 0.00%108 0.786495 422 0.78673 -0.0002 0.0000 0.0134 0.0000 0.00%109 0.852659 90 0.855556 -0.0029 0.0000 0.0124 0.0000 0.00%110 0.924666 13 0.923077 0.0016 0.0000 0.0115 0.0000 0.00%111 1 1 1 0.0000 0.0000 0.0107 0.0000 0.00%

25.7004 13.3667 10.8452 (1) This column is the raw UP-1984 mortality table published by the Society of Actuaries.

85

_______________________________________________________________________________________________ Appendix I Sample 1099-R with Instructions

86

87

88

_______________________________________________________________________________________________ Appendix J Sample Form 5329 with Instructions

89

90

91

92

93

94

_______________________________________________________________________________________________

Glossary AGC Assistant General Counsel’s Office. That department of the Internal Revenue Service generally responsible for the interpretation of IRC §72 and the issuance of related private letter rulings.

AFR Applicable Federal Rate. There are several different AFRs including short-term, mid-term, and long-term. Amortization Method One of three methods permitted for the computation of SEPPs as identified in Notice 89-25. Annuity Method One of three methods permitted for the computation of SEPPs as identified in Notice 89-25. Basis A taxpayer’s lifetime-to-date after-tax contributions to a deferred account. Cumulative Bulletin or C.B. A collection of documents, including Notices, Revenue Rulings, and Revenue Procedures that is published weekly by the IRS in serial fashion. Deferred Accounts Refers collectively to all assets housed inside tax-exempt plans, such as 401(a)s, 401(k)s, 403(b)s, as well as IRA accounts. Distribution Same as a “withdrawal.” Employee Used synonymously with “taxpayer” to indicate the beneficiary of a deferred account. Error, Theory A misinterpretation of a basic concept of law. Error, Practice An administrative or mathematical error in the application of theory. Exhaustion A deferred account that runs out of assets before satisfying the law. IRA Individual Retirement Account, Individual Retirement Arrangement, or Individual Retirement Annuity. IRA, Contributory An IRA to which a taxpayer makes annual contributions that may or may not be tax deductible. IRA, Rollover An IRA that receives the assets from a plan through the use of a “rollover” or “trans-fer.” IRC Internal Revenue Code. IRS Internal Revenue Service, Department of Treasury. LT/AFR Long Term Applicable Federal Rate.

95

Minimum Method One of three methods permitted for the computation of SEPPs as identified in No-tice 89-25. MT/AFR Mid-Term Applicable Federal Rate. Notice A document published by the Internal Revenue Service that should be considered as a primary authority on the subject matter at hand. NUA Net unrealized appreciation (in employer securities). PBGC Pension Benefit Guarantee Corporation. An institution not unlike the FDIC or FSLIC for pen-sion plans. PLR Private Letter Ruling issued by the IRS by request for the use by a single taxpayer. Publication A document published by the Internal Revenue Service that should be considered as a primary authority on the subject matter at hand. QDRO A Qualified Domestic Relations Order is a document issued by a judge pursuant to a divorce proceeding that will govern the splitting of deferred account assets between the ex-spouses. Qualified Plan Any deferred-asset savings plan—defined contribution or defined benefit—qualified under IRC §§401-424. IRAs are specifically excluded. It is the “qualified” designation that causes the tax deferral on assets held within the plan and further provides the plan sponsor (usually an employer) with a tax deduction for any contributions made to the plan. Rollover The method by which an employee can move deferred assets from one location to another. Usually requires the trustee to withhold 20% of the assets as a tax withholding. §72(t) That IRC code section that governs the taxation of withdrawals from deferred accounts. SEP A Simplified Employee Pension. Not to be confused with SEPPs. Separated As in a separation of service from an employer. Can be caused by a voluntary termination, firing, layoff, etc. SEPP A substantially equal periodic payment as defined in §72(t)(2)(A)(iv). SEPP Universe A collection of two or more deferred accounts, segregated from all other deferred ac-counts that are specifically identified as the asset base from which SEPPs will be distributed. SEPPS A substantially equal periodic payment system. Sponsor The enabling entity that creates a deferred plan and trust, usually an employer or collective bargaining unit such as a union. Surtax An additional tax on top of regular taxes; in this case computed as 10% of the amount with-drawn from a deferred account.

96

Table V An addendum to IRS Publication 590 that provides IRS required single-life divisors. Table VI An addendum to IRS Publication 590 that provides IRS required joint- and several-lived di-visors. Taxpayer Used synonymously to mean the employee who is the beneficiary of a deferred account. TCM Tax Court Memorandum. Transfer Usually meaning trustee-to-trustee transfer as a method for an employee to move deferred assets from one location to another. In contrast to a “rollover,” no tax withholding occurs. Trust A separate legal entity as distinguished from an employer/plan sponsor, which actually houses and owns the assets of a plan. Trustee An officer of trust whose job is to safe keep the assets in the trust for the eventual benefit of the beneficiaries. UP-1984 One of several approved mortality tables. This specific one was published by the Society of Actuaries. Withdrawal The removal of assets, usually in cash, from a deferred account.