Joshua Tree Enterprises

Sign Up for Newsletter

About this Entry

This page contains a single entry by lsaret published on September 15, 2008 3:21 AM.

Asset Protection: "UTC Section 505(b) - Pandora Opens The Box" was the previous entry in this blog.

Watching The Watchman is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.

Choices: "Reciprocal Trust Doctrine Threatens Family Decision-making"

Donald Kozusko
Partner, Kozusko Harris Vetter Wareh LLP, Washington, D.C.


The increase in the number and value of family trusts in the United States has been accompanied by an evolution, even revolution, in the ways in which families participate in trust decision-making.  In the past, a single family member might participate as trustee or hold a power of appointment to change the trust disposition, but collective decision-making was rare, even for larger trusts.  A family member might also serve as an "advisor" on investments or distributions, but trust law did not allow a true division of the trustee's legal authority.  The list of choices was short; other alternatives for family participation were either unavailable or unknown.

In contrast, modern trust law and practice have spawned a wide variety of formats and methods for family participation in trust decisions.  Recognizing the reality that not everyone can or should serve alone as an individual trustee, families have utilized these alternatives in their trusts.  Advisors from various disciplines encourage this trend on the ground that it facilitates responsible use and management of wealth. 

But now there are signs that the Internal Revenue Service may undermine this progress by expanding the restrictions it applies to trust decisions it characterizes as tax-sensitive.  Concerned that vote-trading within the family could potentially be used to evade these restrictions, the IRS is considering rules that would effectively ban all family participation in most key trust decisions.  The IRS would apparently justify this ban by combining an expansive reading of the reciprocal trust doctrine with a broader application of its restrictions on powers to change trustees.  The family would then face the threat that its continued involvement would trigger a severe penalty-an estate, gift, or generation-skipping tax on all the assets in all the trusts in question, whether or not there is any evidence of reciprocal voting or even any exercise of the tax-sensitive powers.  Choices for family participation would once again be limited, undermining years of progress in modernizing the role of families in family trusts.

Options abound for opening up decision-making to the family under modern trust law.  The family can take responsibility for its own wealth.  But the efforts of the IRS to combat presumed tax abuse may take away that choice.

Source of the Threat

Under the trust laws of most states today, options abound for opening up the trust structure to the family, so it can collectively participate in the control and management of its wealth.  Trust companies organized under state law but privately owned or controlled by the family or other family trusts--so called "limited purpose," "private," or "family" trust companies--were relatively rare 15 years ago.  Now this option is promoted, or accommodated, by the regulatory law and banking commissions in at least a dozen jurisdictions.  Moreover, for smaller concentrations of wealth, the family can opt for less expensive methods by using multiple family co-trustees, trust committees or specialized fiduciary roles. 


  • Several family members may serve as co-trustees and share sophisticated administrative support staff and systems, with different combinations of individuals serving, as appropriate, to each of the several trusts in question; or different parts of the trustee's role may be allocated among different family members so no one is nominally responsible for everything.

  • Family members can also serve on investment or distribution committees that either make decisions directly, or share decision-making authority with the trustee.

  • Such positions can now be assigned legally meaningful roles if the trust document is governed by the Uniform Trust Act, already enacted in 19 states, or by the laws of Delaware, South Dakota and similar modernized trust codes.  In some states, the family can form an unregulated fiduciary company to hold similar powers.

  • Alternatively, family members can act at an oversight level if they have smaller trusts or less time to devote to these matters.  They can serve on a committee or on the board of a special purpose non-stock company that has various oversight powers, e.g., to change trustees or investment advisors, approve a change to the trust terms, or review the administration of professional trustees.

  • New state trust laws also facilitate amending existing trusts to add choices. 


All of this collective family involvement in family trust affairs--understandably considered a positive development--is now threatened by an unduly restrictive application of the tax law.  The Internal Revenue Service Priority Guidance Plan for 2006-2007 (and the prior year) includes this project:  "Guidance regarding the consequences under various estate, gift, and generation-skipping transfer tax provisions of using a family-owned company as the trustee of a trust."  The IRS is reviewing how decision-making powers held by board members and executives of a family trust company can run afoul of the income-tax grantor trust rules and, more importantly, the estate, gift and generation-skipping rules when a family trust company holds powers that the IRS considers tax-sensitive.

In the simple context of a single trust with an individual trustee who is also a beneficiary, this is a garden variety estate tax issue, e.g., a power to control distributions under a non-ascertainable standard cannot be held by beneficiaries for their personal benefit, even in a trustee capacity, without triggering taxation as a general power of appointment.  The IRS project addresses such questions in the more rarified context of a family-owned trust company.  When a family forms its own trust company, can beneficiaries sit on the distribution committee that would make the same discretionary decisions to distribute to them?  When does that result in transfer tax liability?  Is there any provision for recusal or other prophylactic language that can be used in the trust or trust company documents to avoid such a bad result?

If limited to family trust companies, with no impact on other choices for family participation, the answers to these questions would warrant scant attention by the broader estate planning community.  Unfortunately, death and taxes are not that simple.  Indeed, it appears the IRS analysis is moving in a direction that threatens not only such trust companies, but almost all forms of family participation.

The IRS appears ready to expand the reach of existing precedents and effectively ban family members from exercising tax-sensitive powers, such as distribution powers and powers to change trustees, even when the family is not acting through an elaborate structure and the trust document prevents the affected individual family member from voting on the tax-sensitive question.  For example, in a case where three family co-trustees, A, B, and C, serve together as co-trustees on each of three trusts, and each trust allows for "best interests" discretionary distributions to its respective primary beneficiary (A, B, or C), it would no longer be sufficient to preclude individual A from voting as trustee on a distribution to A, with parallel restrictions on B and C for their trusts.  All of the family trustees would have to be disqualified from voting on "best interests" distributions to any one of them.

Moving in the Wrong Direction

Though Justice Holmes famously declared that the life of the law has not been logic but experience, experience with the life of the Internal Revenue Code is that the IRS, and sometimes Congress, can carry tax logic to the extreme and end up with a result that defies common sense.  In this instance, as a result of the proliferation of state laws permitting family trust companies, the IRS began to receive an increasing number of private letter ruling requests concerning how to treat decision-makers in family trust companies when the trustee's authority included "Tax-Sensitive Powers"--powers held by an individual that would cause  trust property to be included in the power holder's gross estate; a gift made by the power holder to be incomplete for gift tax purposes; or the power holder or another to be treated as the owner of the trust under the grantor trust rules of Subchapter J.

The question in these ruling requests, essentially, was whether including certain kinds of restrictions in the governing documents would insulate the grantors, beneficiaries, and other family members from having to worry about the attribution of Tax-Sensitive Powers held in their capacity as board members, executives, or officers of the trust company.  These requests required extensive document reviews, and while many families did not seek a ruling in these circumstances, the burden was enough to prompt the IRS to seek comments from tax practitioners as to how to replace this process with a revenue ruling or other published guidance.

At the same time, the IRS had begun to associate the family trust company with tax abuse, or at least the potential for tax abuse, as if these trust companies were being formed, or at least could be formed, to make an end run around restrictions that have been applied to choosing an individual to serve as trustee when holding Tax-Sensitive Powers.  Some private letter rulings were explicitly conditioned on very broad prohibitions against family participation in decisions on Tax Sensitive Powers, and others contained such restrictions with sufficient regularity to imply that their presence was considered important to receiving a favorable ruling. 1

The New York Tax Bar seems to endorse the IRS concern that family decision-making in trusts fosters tax abuse.
More recently, the tax bar in New York supported the concerns of the IRS.  In response to the IRS request for comments on how to fashion its new guidance on family trust companies, the New York State Bar Association ("NYSBA") generally endorsed the same broad approach reflected in the private letter rulings.  Specifically, , the NYSBA submitted an extensive discussion and set of recommendations regarding the tax consequences of appointing a family-owned trust company (an "FTC") to serve as trustee of a trust for the benefit of members of the same family (an "FTC Trust"). 2

As explained in the cover letter, the NYSBA Report recommended that the IRS "preclude attribution of Tax Sensitive Powers ... by reason of owning an interest in, being employed by or participating in the governance of a FTC" if the governing documents satisfy the following four requirements. 3

Two of those four requirements provided as follows (emphasis supplied):

(1)     "The by-laws of the FTC should provide that, if a grantor or a beneficiary of an FTC Trust participates in the removal and replacement of a member of a committee who holds powers that are Tax-Sensitive to her, the replacement member must not be related or subordinate to the grantor or such beneficiary,"   (called the "Replacement Restrictions" in the discussion of the Examples below);

(2)     "The by-laws of the FTC should prohibit grantors and beneficiaries from participating in decisions regarding the exercise of any Tax-Sensitive Powers over any trusts as to which any other member of the committee exercising the power is the grantor or a beneficiary.  The purpose of this rule is to avoid reciprocal arrangements among grantors and beneficiaries,"(called the "Participation Restriction" in the discussion of the Examples below).

While these recommended rules on their face apply only to FTCs, the Report takes the general view that these rules employ tax principles that are not so limited and that parallel the accepted practice for individual trustees.  The cover letter concludes:

"Our proposed guidelines are similar to the existing income and transfer tax rules that provide boundaries within which individual and institutional trustees have operated for many years.  We believe that if these rules are incorporated into the proposed guidance, equal tax treatment for all grantors and beneficiaries will be available, regardless of whether an FTC, an independent trust company or an individual trustee is in control."

These broad prohibitions on family participation apparently are being justified by applying the following two principles in combination:

(1)     The "Reciprocal Trust Doctrine." 4  Under this doctrine (also called the "Reciprocal Transfer Doctrine"), interrelated transfers can be "uncrossed" in order to determine who is the transferor in substance of each transfer.  In the typical case, the economic interests or powers granted by person A to or for the benefit of person B can be attributed instead to A if B has made a reciprocal transfer to or for A's benefit, and A's interests and powers under the transfer from B can be attributed to B, and the tax rules applicable to retained interests or Tax Sensitive Powers can be applied accordingly; and

(2)     The principle first adopted in Revenue Ruling 95-58. 5  Under this and related rulings, a power to remove and replace a trustee will cause the trustee's powers to be attributed to the power holder (potentially resulting in the attribution of a Tax Sensitive Power), if the replacement trustee can be a person who is related, or subordinate, to the power holder within the meaning of Internal Revenue Code §672(c). 6

Unfortunately, the position suggested by the IRS private letter rulings and now followed in the NYSBA recommendations does not put proper limits on these two principles and does not acknowledge the volatile impact of combining both rules to justify a new set of prohibitions.

Both restrictions recommended by the New York Tax Bar rest on a single policy judgment: that the sharing of tax-sensitive trust decisions by a family group leads to tax abuse because of the opportunity for collusive behavior, and these powers must therefore be taxed as if they were unqualified and unrestricted.  This represents a new family attribution rule without any legislative authorization or limits.

Boundaries of the Reciprocal Trust Doctrine and Rev. Rul. 95-58
Under existing law and practice, these two principles are limited, or at least have not been extended to certain circumstances.

The Reciprocal Trust Doctrine has not been applied where there has been no reciprocal transfer.  Except one private letter ruling cited in the NYSBA Report [7] 7in which the taxpayer requested the result (and which is inconsistent with a more recent private letter ruling)8the Doctrine has not been applied to uncross interests or powers held by persons who are not transferors, i.e., when the reciprocal powers or interests were conferred by another person.

Similarly, the Reciprocal Trust Doctrine has not been applied to persons holding interests or powers unless there is a showing of interrelated conduct, i.e., the parties have made transfers (or at a minimum engaged in some relevant conduct) in a reciprocal fashion.  Grace and Bischoff, involved demonstrated reciprocal conduct and did not tax assets merely because of the "possibility" of collusion. 9

The power to change trustees in Rev. Rul. 95-58 was "unqualified" (as recognized in the NYSBA Report).  The existing authorities do not directly address, much less resolve, the question of how the power to remove and replace should be treated if the power cannot be exercised without the consent of one or more persons or is otherwise restricted by being shared.

The power to remove and replace the trustee in Rev. Rul. 95-58 was directly exercisable, i.e., it was not attributed through multiple steps, such as a power to remove and replace some other individual fiduciary who in turn had the power (directly, as part of a group, or as an agent of some entity) to make or participate in the decision to change the trustee. 10

If we do not appreciate the significance of these boundaries, and especially if we simultaneously combine the operation of these two principles, the resulting mixture will undoubtedly cause serious practical issues in application, and clearly truncate family participation in trust management.  This can be illustrated by two typical examples involving individual trustees.  As noted earlier, in this discussion we will refer to the first NYSBA recommendation quoted above as the "Replacement Restriction" and the second as the "Participation Restriction."

Example 1

Parent dies leaving two equal trusts for young children, one for the daughter and one for the son, with a family friend as trustee.  Each trust allows discretionary principal distributions without regard to an ascertainable standard.  Some years later the original independent trustee resigns and names the children, now adults, as co-trustees of each trust.  As to each trust, a governing statute prohibits the beneficiary of that trust from exercising a power as trustee to invade principal for personal benefit pursuant to a Tax Sensitive Power, but permits the co-trustee (who is not the beneficiary of that trust, but in this instance is the beneficiary of the similar parallel trust) to exercise the power as if sole trustee, as in NY CLS EPTL § 10-10.1 (2006). (These facts are similar to those in Priv. Ltr. Rul. 9235025, noted above.  Assume that the daughter is not a material contingent beneficiary of the son's trust, and vice versa, so there is no adversity of interests).

Example 1 illustrates the difficulties caused by a jaundiced view of family decision-making, where vote trading is presumed and the opportunity to ask another family member to serve as a member of a committee presumes that you can control the committee's fiduciary decisions:

1(a)      Under the NYSBA analysis the trustee powers in Example 1 would be attributed to the beneficiary-trustee in each trust because of presumed reciprocal conduct, even though the trustees are not transferors.  (Does it matter whether the children were the original co-trustees named by the parent?  What if they were first alternates?)  Priv. Ltr. Rul. 9235025 takes the same position on similar facts, relying on a local law decision that, oddly enough, at the request of the trustees, chose to bless vote-trading by fiduciaries in such a case because the circumstances "invite a trade."  (Question:  What part of trust law sanctions self-interested vote trading?  Answer:  None.)  This earlier PLR is the only support for uncrossing the powers of individual trustees in such a case, but the NYSBA Report also cites the subsequent contrary conclusion in Priv. Ltr. Rul. 9451049 (September 22, 1994).  As to FTCs, the NYSBA Report offers a single rationale for the Participation Restriction:  to prevent "reciprocal arrangements" that would make it "possible" for persons in different family trusts "to work together to ensure that each follows the wishes of the other regarding decisions each could not otherwise make directly".[11] Given this rationale of the perceived need to prevent the "possibility" of collusion, and its position that the rules for FTCs should be the same as for individual trustees, the NYSBA Report seems to adopt Priv. Ltr. Rul. 9235025 as the prevailing law for individual trustees.  

1(b)      If so, it is difficult to explain why the New York statute cited above ignores this result and leaves the trusts exposed to taxation, especially since the statute was clearly intended to curb Tax Sensitive Powers and has actually been amended several times at the request of the New York estate planning bar to achieve certain tax results. 

1(c)      In the same way, the NYSBA Report recommends the Participation Restriction for FTCs and states that the recommendations would achieve "equal tax treatment" for all trustee arrangements.  The Participation Restriction applies a taxable taint merely because the participant has a trust subject to the decision of other members of the committee.  That analysis would also threaten a taxable result in the case of individual trustees who are related, or even for those who serve as trustees of each other's trusts because they are neighbors, business associates, or in the best foursome at the country club, since abusive vote-trading can be conclusively presumed under this view.

1(d)      These two rules (the Participation Restriction and the Replacement Restriction) are even more problematic because they could both apply together.  For example, do the facts in Example 1 always lead to attribution of the power if the children became co-trustees by exercising a shared power to remove and replace the original trustee and thereby appointed themselves as co-trustees? 

1(e)      Restrictions on "participation" naturally raise the question of how to define that term.  Voting, or mere "participation" in the discussion?  Presence in the room?  A ban on any involvement in the process would certainly be an odd requirement to impose on a beneficiary.  Vague terminology is not a suitable foundation for administering a tax of over 40% of asset value.

Example 2

While in their sixties, husband and wife each simultaneously establish funded revocable trusts as the core document in their respective estate plans, providing after death for a sprinkling trust, without an ascertainable standard, for the children and surviving spouse in each case, and after the surviving spouse's death, for the sprinkling trust to divide into two trusts, one for each child.  All of the family members are co-trustees.  Assume local trust law applies the same rules as in the New York statute in Example 1 and that the trust agreement provides those co-trustees permitted to exercise a power may act by majority vote (or unanimously if there are only two co-trustees).  Husband and wife die 10 years apart without changing these documents.

2(a)      Does existing law and practice completely preclude the surviving spouse's participation as trustee on distribution decisions for the sprinkling trust?  Presumably not, since the two spouses obviously cannot act as trustees over each other's sprinkling trust.  Therefore, it is very hard to conclude that the trusts are reciprocal in any sense.  The children had even less opportunity to collaborate on determining the final trust terms, so the trusts should not be considered reciprocal as to them under existing law and practice.  The trusts are simply not interrelated and the trustees were not transferors of any contemporaneous trusts.

2(b)      Yet, the vote-trading analysis of the NYSBA Report would find taxable reciprocity here simply because the trustees are family members.  For example, if the family in Example 2 later chose to establish an FTC, none of the three surviving family members could serve on the trust company's distribution committee under the NYSBA's recommended Participation Restriction, which is designed to preclude vote-trading.  Without the Participation Restriction, the vote-trading analysis of the NYSBA Report would cross-attribute powers, causing taxation unless the exception to Internal Revenue Code § 2041(b) applies (if the trustee powers are not Tax Sensitive because the trustees are adverse to each other as to the exercise of the power).[12]

2(c)      This vote trading analysis has boundless potential to cause mischief, and taxation.  Going back to the original Example 2 with individual trustees, assume in Example 2 that the wife is the surviving spouse and she is also the trustee of an irrevocable gift trust funded by the husband many years ago, established for the benefit of the children when they were first born.  No FTC is involved, but the vote-trading analysis of the NYSBA presumes she will trade distributions from the gift trust to buy her children's votes over the sprinkling trust.  Even if the gift trust has an ascertainable standard, the conclusion would hold; that is, if the ascertainable standard is written as a limit on distributions rather than as requiring distributions for support, her veto power still provides leverage for trading votes.  Yet the traditional Reciprocal Trust Doctrine should not apply because the two trusts are not interrelated at all, either in time, by parallel transferors, or by similarity of terms.  A jaundiced view of family decision-making has no limits, but the Reciprocal Trust Doctrine does.

2(d)      If this negative view of family decision-making is adopted, it seems likely that the "adverse interest" exception will rarely prevent taxation because, under this analysis, powers can be attributed across a large group of participants even when the ability of any one of them to influence the result is marginal and they are not beneficiaries of each other's trusts.  Put another way, if vote-trading can be conclusively presumed, the analysis of what is an "adverse interest" becomes elastic and uncontrolled in most family participation structures.  Adversity can be presumed to vanish because vote trading can be used to make an "end run" around adverse interests.  For example, in 2(b) above, there would be no adverse interest exception because the surviving spouse can--and presumably will, according to the theory--reciprocally trade votes with one of the children in order to deplete the trust in favor of those two family members.  This theory has no discernible limits once we presume that the members of a family will trade votes even when they have not collaborated in the first instance as transferors to create interrelated powers and interests.
2(e)      Going back to the original Example 2 with individual trustees, what would be the result under existing law and practice if the surviving spouse had the power to remove and replace one of the children as trustee, with the consent of the other child, and could then appoint a related party as a replacement?  The power is shared and thus not "unqualified" as in Rev. Rul. 95-58.  Yet, the Replacement Restriction is characterized in the NYSBA Report as treating FTCs in a manner equivalent to existing law for individual trustees, and clearly applies to shared powers, i.e., where the beneficiary merely "participates" in the replacement decision.

2(f)       Assume, in Example 2, a family company or family committee holds the power to change individual trustees, and the three family members in that Example serve as the decision-makers.  If the Replacement Restriction recommended by the NYSBA applies to FTCs it should apply to this case to prevent, for example, the appointment of a replacement trustee who is a spouse of any one or more of the three family members.  The naming of spouses would be treated as reciprocal conduct and the spouse of one member treated as "related" to any other member of the group making the appointment.  Note the Replacement Restriction applies to appointments of persons related to the grantor or beneficiary in question, not persons related to any member in the group, but the use of the term "Tax Sensitive Power" in the NYSBA Report appears to include a power to change trustees and thus to broaden the application of the rule.  The Report also employs an illustration suggesting the Participation Restriction is designed to prevent "abuse through such reciprocal arrangements" by which two individuals holding replacement powers can replace members of the distribution committee with related parties by arranging for each to exercise the power with respect to the other's trust.[13]


A ban on family participation would be virtually impossible to articulate and administer without endless line drawing and questionable judgments, which then undercuts the purpose of the IRS guidance in providing clarity for family trust companies.  Moreover, if adopted, such rules would cast considerable doubt on the tax treatment of more common trust structures, in the absence of a consistent and defensible explanation.

Some families would inadvertently over the course of time fall into this trap of presumed abuse.  Other families who actively pursue family decision-making in its various modern forms would be discouraged by the need for extensive legal analysis and the continuing threat of taxation.  These difficulties can be avoided by confining Rev. Rul. 95-58 and the Reciprocal Trust Doctrine to their existing limits and adopting a policy that does not presume tax abuse when a family participates in trust decisions.

1 See, e.g., Priv. Ltr. Rul. 200410015; Priv. Ltr. Rul. 200546052.  See Donald D. Kozusko and Miles C. Padgett, Private Trust and Protector Companies: How Much Family Control? 43 Tax Man. Memo. No. 22 (Nov. 4, 2002).

2NYSBA Tax Section Report (No. 1111), Tax Consequences of Appointing Family-Owned Trust Companies, June 1, 2006.

3 Id.

4See United States v. Grace, 395 U.S. 316 (1969); Estate of Bischoff v. Commissioner, 69 T.C. 32 (1977).

51995-2 CB 191.

6 IRC § 672(c).

7 Priv. Ltr. Rul. 9230025.

8 Priv. Ltr. Rul. 9451049 (September 22, 1994)

9 See Paul E. Van Horn, Revisiting the Reciprocal Trust Doctrine, 30 Tax Man. Estates Gifts and Trusts Journal  Memo. 224 (July 14, 2005).   

10 Indeed Rev. Rul 95-58 itself goes beyond any positions sustained by the courts. See Comments to Office of Chief Counsel on Family-Owned Trust Companies by McGuire Woods, LLP, Feb. 27, 2006; Donald D. Kozusko and Miles C. Padgett, Private Trust and Protector Companies: How Much Family Control?43 Tax Man. Memo. No. 22 (Nov. 4, 2002).

11 NYSBA Report, at IV.C.

12 IRC § 2041(b). 

13 NYSBA Report, at IV.C.