Sec. 2056(b)(5) allows a marital deduction where a life estate is created for the benefit of the surviving spouse, with a power of appointment over the entire interest solely exercisable by the surviving spouse. Since the surviving spouse was entitled to all of the income, which would be distributed at least quarterly, and only the surviving spouse had the power to appoint the principal of the pension trust, the IRS concluded that the surviving spouse's interest in the pension plan qualified for the marital deduction. The marital deduction is also available to property placed in qualified terminable interest property trusts (Q-TIP) under Sec. 2056(b)(7).

Rev. Rul 89-89, 1999-27 I.R.B. 11 involved installment distributions from an IRA to a Sec. 2056(b)(7) Q-TIP trust. The IRS ruled that the IRA itself qualified as a Q-TIP trust since the income earned on the undistributed portion of the IRA, as well as the income earned on the portion of the IRA distributed to the Q-TIP trust, would be distributed annually to the decedent's spouse for life and no person other than the spouse would have a power of appointment.

In Rev. Rul. 89-89, the decedent established an IRA account and designated the trustee of a testamentary trust as the beneficiary in the event of his death. The trust provided that all income was to be distributed at least annually to the decedent's surviving spouse and no other person was entitled to appoint trust principal to anyone other than the surviving spouse. As such, the trust qualified as a Q-TIP trust under Sec. 2056(b)(7) and the Ruling indicates that the executor intended to so elect.

Pursuant to the terms of the IRA, upon the decedent's death, the IRA will distribute a sufficient amount of principal to the Q-TIP trust, based on the life expectancy of the surviving spouse, in order to meet the requirements of Sec. 401(a)(9). In addition, the IRA will also distribute any income earned during the year on the undistributed portion of the IRA to the testamentary trust, which in turn will pass through the income to the surviving spouse. Upon the surviving spouse's death, the balance of the IRA is to be distributed to the Q-TIP trust.

The IRS ruled that the IRA will qualify for Q-TIP treatment and, thus, the decedent's estate will be entitled to claim a marital deduction for the value of the IRA. The IRS reasoned that the Q-TIP trust acts as a mere conduit so that all income earned by the IRA will, in fact, be distributed to the surviving spouse. As a result, any distribution of income from the IRA to the Q-TIP trust will be deemed made directly to the surviving spouse.

Post Death Events Impact on Deductibility of Expenses

Claims which are potential, contingent, contested, or unmatured at a decedent's date of death require analysis of post death events in order to determine deductibility on a decedent's estate tax return. In Estate of Charles P. Cafaro, Deceased, T.C. Memo 1989-348 (1989), the Tax Court applied this rule to the alleged debts of the decedent.

The decedent died in 1978. The Court considered three debts deducted on the estate tax return. The first related to a stock purchase contract. The executrix deducted $4000 on the estate tax return, though the creditor in 1984 accepted $2000 in full satisfaction of the claim.

The second debt was for advertising services deducted on the estate return in the amount of $22,000. This claim was settled for $15,000 in 1980.

The third debt related to a bank loan.

The decedent had signed certain loan agreements not only as president of the debtor corporation but also in his individual capacity. It was not clear whether he was signing as comaker or guarantor. Decedent's estate never was required to make any payments on the bank loan because they were made by the primary obligor.

IRS disallowed deductions for the bank loans and limited the other debts to the amounts paid. The estate argued that post death events were irrelevant in considering the deductibility or enforceability of claims.

The Court, agreeing with the IRS, found no evidence that the contract debt or advertising debt was for a sum, certain or legally enforceable on the decedent's date of death.

The Court ruled that, with regard to the bank loans, since the decedent had not received any proceeds of the note, the decedent was a guarantor, not a co-maker, as the principal debtor was not in default as of the decedent's date of death. The claims were unmatured and merely contingent and potential. The Court held that post death events had to be considered in determining the enforceability and value of all three claims. The amount of deduction allowed by the Court was limited to the amount eventually paid by the estate to settle the claims.

Succession and Estate Planning for S Corporation Owners

S corporations have increased in popularity as a result of TRA 86 and the lowering of personal income tax rates. More corporations are electing S corporation status and therefore, the transfer of S corporation stock to trust instruments has become more important to the estate plan of business owners. This article discusses the types of trust vehicles which can currently own S corporation stock, the inherent problems, and some ways to add flexibility in their use.

Only certain irrevocable trusts may own S corporation stock and not cause an S election to terminate. One is a grantor trust, under which an individual U.S. citizen or resident is taxed on all of the domestic trust's income under Secs. 671 through 678 of Subpart E. In this type of trust, the activities of the trust are taxed to the "grantor" for all items of income, deductions and credits. The other type that will qualify as a permissible shareholder is a qualified subchapter S trust (QSST), which is a domestic trust that meets all of the following requirements under Sec. 1361(d):

1. There is only one current income beneficiary;

2. All of the income of the trust is distributed or required to be distributed currently to that income beneficiary;

3. The income beneficiary is the only person who may receive distribution of trust principal during his or her life;

4. The income interest in the trust must terminate on the earlier of that beneficiary's death or the termination of the trust; and

5. Upon termination of the trust during the beneficiary's life, the trust must distribute all of its assets to the income beneficiary.

The beneficiary of the trust or that beneficiary's legal representative must file an election to be treated as a QSST. A separate election is made with respect to each S corporation.

QSSTs Are Not Flexible

The problem posed by the QSST is the lack of flexibility. The owner of the shares of S corporation stock has no ability to make gifts into a single trust for the benefit of all of the children.

An alternate structure which avoids some of the limitations of a QSST is to use a Crummy withdrawal power to create a grantor trust under Sec. 678. According to Sec. 678, upon the lapse of the withdrawal power, a person other than the creator can be treated as the owner of a trust. As a result, income can be accumulated.

Under a Crummy power, a specific party or powerholder is given a right to withdraw a portion of the assets of the trust. The lapse of the power occurs when that individual does not elect to exercise. The use of such powers has historically been to obtain present interest gift tax exclusion for gifts to irrevocable trusts. When the withdrawal powers lapse and Sec. 678(a) applies, as in PLR 8521060, the power holder can be treated as if he had released the power to withdraw, and therefore be considered the owner of the trust.

In recent PLRs 8805032 and 8809043, the IRS applied similar logic directly to S corporation situations. They concluded that the failure to exercise a withdrawal power causes the powerholder to be treated as the owner of that portion of the trust. If the powerholder had held such lapsed withdrawal powers over the entire corpus of the trust, then the trust will qualify under Sec. 1361 (c)(2)(A)(1) as an eligible shareholder of an S corporation. The S corporation election will not be revoked.

Succession Problems Not Solved

A major benefit of this structure, according to the rulings, is that income can be accumulated for later distribution. However, it does not solve long-range succession planning problems such as which individuals should ultimately own the shares of stock, nor does it address the potential gift and estate tax problems associated with the beneficiaries.

Two potential solutions exist with regard to succession planning. First, an irrevocable trust with a Crummy withdrawal power can be designed which runs for the life of the sole beneficiary and which establishes a testamentary limited power of appointment by the settlor's spouse. Upon the termination of the trust, corpus can be distributed to the beneficiary and other descendants of the original "grantor." Unfortunately, this alternative may create serious problems under the Spousal Unity Rule of Sec. 2036(c)(3)C. A probably less risky approach is to create a trust for each child which terminates when the youngest one attains a specific age. At that time, corpus can be distributed equally among the grantor children who are full time employees of the corporation. This provides an effective way to ensure that the stock passes to the children who are working in the business.

When establishing an irrevocable trust with a Crummy power, two important considerations are the gift and estate tax consequences. A lapse of the withdrawal power for amounts in excess of $5,000 or 5% of the trust principal will be considered a transfer for purposes of Sec. 2036 and so most, if not all, of the trust will be included in the beneficiaries' estate for purposes of federal estate tax. In addition, when the lapse of a power exceeds $5,000 or 5%, under Sec. 2514(c) a gift will be deemed to have been made by the beneficiary. Such a gift would probably be considered a future interest, not qualifying for the $10,000 annual exclusion under Sec. 2503.

A more sound approach which will allow the maximum amount of value to pass free of gift and estate taxes is to establish, at the inception of the corporation, the irrevocable trusts with Crummy powers which will qualify as S corporation shareholders. The grantor can gift stock currently valued under $5,000 with significant appreciation value, with no potential gift or estate tax problems upon the lapse of the withdrawal powers.

Selecting Property for Estate Thinning

Once the desirability of reducing (thinning) a client's estate has been determined, estate planners must select the best properties for thinning. Capital gains tax liability makes a potentially simple problem more difficult because property given away will not gain the step-up in basis available to property retained. Confronted with a portfolio of properties possessing various combinations of unrealized capital gains, current value, and potential future appreciation, a planner must determine which ones are better to keep and which should be given away.

The estate thinning goal is to minimize taxes, not maximize return on investment. The latter goal has already been considered in the initial selection and subsequent retention of the properties in the client's portfolio. Since estate thinning merely shifts assets within a family unit, return on investment is not affected by the thinning decision, considering the family as a unit. Total tax liability, on the other hand, is affected.

Of the four possible taxes--capital gains, income, gift, and estate-- only the first and last are relevant to the decision-making process. Income tax is irrelevant, because the level of income tax liability has already been decided in the initial selection of investments. In addition, since the donor and donee(s) are most likely in the same tax bracket, income tax liability is not affected by the thinning decision. While this assumption may not hold in every case, it is generally true. Gift tax is irrelevant because a transfer resulting in a taxable gift is not dependent on which property is selected for thinning. All that matters is the current value of the gift. On the contrary, capital gains tax is relevant since the tax liability differs depending on whether a property is held in the client's estate or divested. While it is not certain that property will be sold at death, it is a possibility. Estate planners will want to minimize the tax liability. Hence, it is important to consider capital gains tax liability, understanding that the thinning decision must be based upon a comparison of a sale of property at death of a client under two options: with the property both in and outside the client's estate. Finally, estate tax is relevant because, even though the current value of a gift may be the same regardless of which properties constitute the gift, the future values of those properties are likely to differ and, hence, will have an impact on the estate tax liability.

In considering which properties to select for thinning, the relevant factor is the anticipated size of capital gains of various properties at the client's death. This is a function of the current unrealized capital gain and expected appreciation. The latter in turn is a function of the rate and length of time of appreciation. Thus, in sum, the relevant factors are future market value and tax basis.

The $64 Question

The planner must ask this question: Among properties in a portfolio, which one(s) should be divested rather than held so as to minimize the sum of capital gains and estate taxes? The solution thus involves comparing various alternatives.

Some situations present easy solutions. Among properties with the same estate tax liability but differing capital gains tax liabilities, those with the smallest capital gains tax liability should be divested and those with the largest retained, thereby taking maximum advantage of the step-up in basis at death. Alternatively, among properties with the same capital gains tax liability but differing estate tax liabilities, those with the lowest estate tax liability should be retained and those with the highest divested.

The more difficult situations, however, are those where the estate and capital gains tax liabilities differ among properties. For example, consider two properties, each currently worth $10,000, the first with a basis of $6,000 and an expected future value of $13,000 and the second with a basis of $1,000 and a future value of $15,000. Note that the eventual capital gain will be $7,000 on the first property and $14,000 on the second. Which property should be held and which divested? If the first property is held, estate tax is minimized, but capital gains tax is high. If the second property is held, capital gains tax is lower, but the estate tax is higher.

Which Strategy is Better?

Conceptually, the answer is simple. Compare the total tax liabilities of the two strategies (hold No. 1 and divest No. 2 versus divest No. 1 and hold No. 2) and select the strategy with the lower total tax liability. The problem, however, arises when there are many properties to be evaluated. In a portfolio of merely six properties there are fifteen possible pairs. The solution involves using a simple spreadsheet. Future market value and capital gains of each property are entered. The spreadsheet automatically calculates the estate tax for each property, were it to be held, and the capital gains tax, were it to be divested, as well as the difference between the two. The planner ranks the assets according to the magnitude of this final number, estate tax minus capital gains tax. Those properties with the lowest values are most desirable to hold, and those with the highest are most desirable to divest.

The problem with this solution is evident: How does the planner estimate future market value? The variables, as indicated above, are the rate and time of appreciation; both must be assumed. As a practical matter, however, time is not a factor. The time frame ends at the same date for all properties, the date of death. Rate of appreciation is the major variable, but only if large differences are assumed. Therefore, as long as the planning horizon does not exceed about five years, the model does not appear to be significantly sensitive to rate and time assumptions.

An estate planner may wish to include estate administration expenses as a factor. If so, an expense rate may simply be added to the assumed estate tax rate.

Using this solution, the estate planner replaces a qualitative decision with a quantitative model. Clients will be able to grasp the consequences of the various alternatives, but should be cautioned that future values are subject to change from those used in the model.

Trapping Distribution Affects Characterization of Income Distributed

A decision handed down by the Tax Court in late 1987 dealth with the characterization of trust income from a trapping distribution. Specifically, the court deliberated whether the character (as between taxable and tax-exempt income) of amounts reportable by the beneficiary of a simple trust is determined solely by the trust's internally generated income, or should the amounts received by the trust in a trapping distribution from an estate also be considered. The Court ruled in favor of the Commissioner holding that the trapping distribution was improperly ignored by the petitioner.

The petitioner was the income beneficiary of a testamentary trust created under the will of her husband. By its terms, the trust was a simple trust requiring the trustee to distribute all net income to the petitioner at least annually for her lifetime. In 1980, the estate of petitioner's husband generated distributable net income (DNI). The estate then made a principal distribution to the testamentary trust--a trapping distribution--causing a portion of the estate's DNI to be taxed to the trust. For tax purposes, the estate principal distribution constituted income to the trust and was part of the trust's DNI. However, for fiduciary accounting purposes, the distribution retained its character as principal. Therefore, the taxable income generated by the estate distribution was "trapped" at the trust level.

When calculating the DNI of the trust and characterizing the composition of DNI, the petitioner did not include the trapping distribution.

Trapping Distribution as Part of DNI

Under petitioner's calculation, a large portion of DNI was attributed to tax-exempt income--thus, not taxable to petitioner as beneficiary. The Commissioner maintained in his notice of deficiency that the petitioner must include the trapping distribution as part of the trust's DNI when determining the amount and character of the income attributable to her interest in the trust.

Under both the petitioner's and Commissioner's calculations, the petitioner would receive the same amount of income, since the amount of accounting income is the basis for distribution. The Commissioner, however, changed the character of income passing to the petitioner by including the trapping distribution in DNI. The Commissioner applied a ratio to the DNI, the numerator of which was gross internally generated income, and the denominator was trust DNI. The calculation increased the proportion of taxable income attributable to petitioner's interest in the trust.

This case illustrates the paradox that arises in this situation. The income received from the estate and the internally generated income both go into the computation of the trust's DNI. However, under local law, the trapping distribution is not distributed to the beneficiary and remains part of trust corpus. What results is "undistributable distributable net income." However, the trust is governed by the trust instrument which specifies that only the net accounting income be distributed.

The Court was not persuaded by the petitioner's argument under Reg. 1.652(b)-2 that particular classes of income which constitute trust DNI be allocated to the persons entitled to receive the principal or income of the trust. The Court stated that petitioner has incorrectly equated "principal" and "income" with "different classes of income" under Sec. 652. Principal and income are terms used under the trust instrument as defined by state law and do not address differentiating among classes of taxable income. On the other hand, the designation of different classes of income under Sec. 652 (dividends, tax-exempt interest and capital gains) has a tax impact on the beneficiary. Neither the trust instrument nor local law specifically allocated different classes of income to the petitioner. Therefore, DNI had to be allocated between the petitioner and the trust proportionately.

Tracing Does Not Apply

The Court reviewed the history of the DNI concept which eliminated any tracing requirements by presuming that a distribution is made out of the trust's DNI regardless of whether the distribution was actually made from income or principal. Tracing is only permitted in limited cases where the trust instrument or local law specifically requires a nonproportionate allocation of classes of income among beneficiaries.

In sum, since the trust instrument and local law are silent as to specific allocations of different classes of income to different beneficiaries, petitioner must include in income the same proportion of each class of income constituting the DNI of the trust as the total of each class of income bears to the total DNI of the trust. Therefore, the Commissioner's position and computations were upheld.

Gabe M. Wolosky, JD, LLM, CPA, Prager and Fenton Dan A. Diers, CPA, Rashba & Pokart David Vigoda, CFP, CFA with assistance by Philip J. Vecchio, JD, CPA, Bollam, Sheedy, Torani & Co. Darryl Eve Marschke, Esq., Ernst & Young