TRUSTS FOR THE NEW CENTURY
THE THIRD PARADIGM

Mark B. Edwards

Of Counsel
Poyner & Spruill, L.L.P.
Charlotte, NC 28202

©Mark B. Edwards 1999 (Version 101599)

OUTLINE OF TOPIC

1. Introduction
2. How We Got Where We Are

1. The First Paradigm – The Real Estate Model
2. The Second Paradigm – The Tax Model

1. Why Do We Have To Embrace A New Paradigm

1. The Change In The Nature Of Wealth
2. The Change In The Management Of Wealth
3. The Change in the Taxation of Wealth
4. The Change In The Manager Of Wealth

1. Dealing With The Challenge Of Total Return Investing

1. Rely Upon The New Uniform Principal And Income Act
2. Give Discretionary Powers To Trustee
3. Use Unitrust Payment Formula

1. The Private Unitrust – Design Considerations

1. The Distribution Amount In The Ordinary Unitrust
2. A Variation – The Percentage Unitrust

1. The Mandatory Percentage Unitrust
2. The Withdrawal Percentage Unitrust

1. Qualification For The Marital Deduction
2. Additional Discretionary Distributions
3. Source Of Distributions

1. The Other Challenge – Providing A Trustee

1. What Is The Problem
2. A Statement of the Purpose
3. The Power To Remove And To Replace The Trustee
4. The Trust Protector ( Or Special Trustee Or Trust Advisor}
5. Who Can Be The Trust Protector
6. A Word Of Caution

1. Introduction.

What estate planners do today is largely centered on using trusts to reduce wealth-transfer taxes. And we do a fine job of it; the trusts we draft are carefully structured to comply with all requirements of the Internal Revenue Code and the regulations interpreting it, while at the same time meeting all those rules about trust law that we learned in law school.

In most of these trusts the distribution pattern is the same, save for minor variations:

The trustee shall hold the trust corpus and pay the income therefrom to A during A’s lifetime. Upon A’s death, the trustee shall divide the trust into two equal shares and pay over one share to B and one share to C.

True, in some instances the trustee may be given discretionary power to pay income or principal to one or more beneficiaries. But even so the common pattern is someone receiving something now, and a different or additional someone receiving what is left later.

For the 35 years of my full-time law practice, I drafted trusts this way. Then I left the practice to be a private trustee, serving under documents that, for the most part, I had prepared. Chefs will tell you that you never enjoy your own cooking. How true that is! I find that I do not enjoy serving under some of the documents I so proudly drafted. Not that they are wrong in any legal sense, mind you. Not at all. But they do not, I now believe, respond properly to the need of the trustee and the beneficiary in today’s financial environment.

The thesis of this paper, based upon the experience of "eating my own cooking," is that we must change our way of drafting these trusts in the future. We must have a new model, a new paradigm, around which to fashion the provisions of the trusts we draw. We must spend more time and creative energy developing trust operating plans which comply with the requirements of the tax law. We must permit the trust to invest its assets in a manner consistent with modern investment practices and modern trust law, and enable the trust to meet the human needs and desires of both the grantor and the beneficiary.

2. How We Got Where We Are.

Before we look at a new paradigm for our trusts, it will be helpful to see what has brought us to this place in the history of the trust.

A. The First Paradigm - The Real Estate Model

Long ago, in medieval England, wealth and land were synonymous. The nobles controlled all land, and used it to buy or reward lesser nobles who assisted them. Upon the death of these lesser nobles, the land reverted to its original owner who then used it to replace the deceased individual. To avoid the loss of land upon death and to avoid some of the other incidents of the feudal system, some bright conveyancer of land conceived of transferring land to one or more persons "for the use of another." Since the multiple holders of the use never completely died, there was no reversion, and since the land user did not "own" the land, no feudal duties were owed.

Uses became very popular and, according to one estimate, by the early 1400’s, the majority of real estate in England was held "for the use of " someone other than the title holder. This removal of the land from the feudal system became intolerable for the Crown, so two of our favorite rules of law – The Statute of Uses and the Rule Against Perpetuities – were developed to curtail uses.

Out of this conflict, however, came the equitable concept of the active trust, which was the forerunner of the trusts we use today. This first kind of trust was conceived of and implemented as a way to hold title of real estate in one person for the use or benefit of another. This was the First Paradigm.

B. The Second Paradigm – The Tax Model

The Second Paradigm originated in 1916 when the estate tax was reintroduced to help finance World War I, and it has blossomed ever since.. It is a new view of the trust as a device primarily for tax purposes, to reduce wealth transfer tax, to manage income tax, and to avoid generation-skipping tax. This tax-centered Second Paradigm is added to and melded with the First Paradigm of the trust as a property-owning, probate-avoiding way to own property.

The focus of the Second Paradigm is on the act of severing the title of the property from the client and transferring it to the trust in a manner that excludes the property from the transferor’s estate and sometimes even from the estate of the beneficiary. We focus on the creation of a trust to reduce taxes at the moment of the grantor’s death. And that is where we stop! We give little thought to how that trust will operate for the next 25 years. We simply say to the trustee and to the beneficiaries: "Bon Voyage! Have a nice life with the tax money I have saved for you."

But now we need the Third Paradigm for the trust, a paradigm that takes the elements of the first two models and adds to them a forward-looking concern for the operation of the trust after its execution and funding. This new paradigm will focus on implementing the client’s wishes with respect to the enjoyment and management of the property that remains after the tax is paid, rather than the tax consequences at the moment of the grantor’s death. It will be a trust driven by investments, not taxes!

3. Why Do We Have to Embrace a New Paradigm?

The move from the First Paradigm to the Second was caused by the imposition of particular taxes – wealth transfer taxes – by the government in 1916 to finance the First World War. This event caused a change in the way trusts were viewed. The paradigm shifted!

Now it is again external forces that require us to rethink the way we look at and deal with trusts, and compel us to move to the Third Paradigm. These forces are broad currents within our society, and they are four in number – the change in the nature of wealth, the change in the management of wealth, the change in the taxation of wealth, and the change in the manager of wealth.

1. The Change in the Nature of Wealth

During the formative years of the trust concept, wealth was represented by land. The fruit of the land – the crops – could be consumed or traded away while the land remained to produce another crop. The land was "principal," and the crops were "income." "Income" was disposable; "principal" was not. Today, however, the nature of wealth has changed; wealth is now represented primarily by financial instruments. Crop rotation has become asset allocation!

Because of the long-term influence of the First Paradigm, even as the nature of wealth changed, we comfortably moved to the concept that these financial assets could be "rented" just like farm land. That is, we could rent our money to Duke Energy, for example, by buying some of its shares. In return, we receive a "crop" of dividends. We call the value of the shares "principal," and we call the dividends "income." We consider these two elements as two distinct things just as the trustee in the First Paradigm did. But in doing so we have ignored that inescapable fact that there is a qualitative difference between the land and the crops of the First Paradigm. Land and wheat are different.

In the world of financial assets, however, there exists no such difference. As one commentator has so pithily noted:

Property is property. Money is fungible. Accretions to wealth derived by the owner from cash payments made by the issuers of securities (interest and dividends) have no essential characteristic that distinguishes them from those derived from cash payments made by the purchasers of securities sold by the owner (realized capital appreciation). And, absent the antiquated model and associated slogans, no one in her/his right mind would ever have thought otherwise.

In other words, since wealth is now represented largely by financial capital rather than real property, we must move away from the artificial distinction between income and principal found in the First and Second Paradigms. We must move to the Third Paradigm.

2. The Change in the Management of Wealth

The change in the nature of wealth has lead to a change in the management of wealth, and this new form of wealth management is called "total return investing." This simply means the investing of funds for maximum return, regardless of whether the return is in the form of accounting income or capital appreciation. This form of wealth management has been accepted by the financial community for several reasons:

• The seemingly unending increase in the price of equity securities;

• The disappearance of meaningful yields on those same securities; and

• The advent of ERISA.

In 1974, Congress passed the Employee Retirement Income Security Act (ERISA). This Act and subsequent follow-up statutes have created a multitude of accounts for individuals under profit sharing plans, 401(k) plans and IRAs. All of these accounts share a common trait: no distinction is made between "income" and "principal." Because of this, the account holder does not track his income or her capital gains. Instead he or she simply looks at the account and compares its total value today to the total value at the start of the period. In other words, they have embraced total return investing.

In recognition of this movement, fiduciary law is changing (or perhaps I should say, has changed) to embrace total return investing as the standard for fiduciaries. The logic of this adoption goes like this. A basic tenet of trust law is that a trustee must deal with all beneficiaries of the trust impartially. However, the artificial distinction between principal and income inherently favors one class of beneficiaries – either the current income beneficiaries or the remainder men – over the other. Therefore, both the Uniform Prudent Investor Act and the Restatement (Third) of Trusts have explicitly adopted total return investing as the standard for proper fiduciary conduct.

But if trusts continue to be drawn in a way that divides income and principal into two distinct piles of money and provides a separate treatment for each pile, a problem is created for the trustee. Let me summarize this investment problem created by the First Paradigm mode of drafting trusts:

• The trustee must act with impartiality considering the needs of the current beneficiaries and the anticipated needs of the future ones.

• Attorneys draw trusts requiring all income to be distributed to one set of beneficiaries and the principal to be held for another.

• The higher the needs of the current beneficiaries, the greater amount of the trust assets must be invested in income-oriented securities, and these are the investments that are least able to meet the needs of the future beneficiaries

• The beneficiaries are thus disappointed with impartiality. The income beneficiaries feel that the income produced is inadequate, while the remaindermen feel the trustee has not produced sufficient growth.

This investment dilemma and the new fiduciary standards lead, I believe, to the conclusion that trustees must move to the new way of total return investing and that the trust documents under which they serve must facilitate, not impede, this new investment methodology. In short, there must be a move to the Third Paradigm.

1. The Change (?) in the Taxation of Wealth

The Second Paradigm for trusts was begun in 1916 with the permanent enactment of wealth transfer taxes. With the estate and gift taxes now reaching a marginal rate of 55% at the relatively low level of $3,000,000 and with a generation-skipping tax of 55% applying to many typical trust dispositions, the defining force in trust drafting became the desire to minimize these taxes.

But what happens if this defining force disappears? The tax bill recently passed by Congress contains provisions which would ultimately eliminate the wealth transfer taxes by 2009. While the President vetoed this legislation, it does seem that Congress is serious about doing something to lessen the burdens caused by these levies. Pundits surmise that at the least we are likely to see an increase on the applicable exclusion amount to, say, $1,500,000 per person and perhaps an increase in the annual exclusion amount under §2503 to $20,000. Changes of this magnitude would remove wealth taxes as an estate planning consideration for many people. Of course, there will always be persons with more, but they become fewer and fewer as the exempt amount increases. We cannot sell tax-driven trusts to the multitude of these people, but investment-driven trusts would be of interest to all.

2. The Change in the Manager of the Wealth

Even as the changes in the nature, management and taxation of wealth have been occurring around us, another change has been taking place in the world of trusts. Every trust must have a trustee, and that trustee must be sophisticated in fiduciary matters, disinterested in the financial results of the trust and sufficiently long-lived and stable to render the necessary services for the life of the trust. In other words, the trustee should be a professional trustee. And this normally means a corporate trustee!

Throughout the early years of my law practice, the use of corporate trustees was very common. In recent years, however, there has been increasing reluctance on the part of many of my clients to utilize their service. Part of this reluctance, I believe, relates to the unfulfilled expectations of many trust beneficiaries, for which our trust documents bear some responsibility. Tales of these disappointments, told across the table at the Wednesday bridge club or around the lockers at the golf course, have been a powerful, negative influence on my clients and their willingness to use a corporate fiduciary.

Part of the reluctance, however, is brought about by the banks themselves. Banks are bigger and more impersonal. Customers interact with them by use of ATMs, toll free phone numbers and the Internet. The image of the banker as a personal financial advisor has faded away. Often in today’s bank, there is no "Trust Department;" banks now have their "Capital Management Division." The name presumably reflects the priorities of the institution! The primary mission of the corporate trustee as defined by their literature, their advertising and their presentations seems to be to provide asset management, not to offer wise counsel to sometimes immature and argumentative beneficiaries nor to exercise considered discretion in difficult circumstances.

We must also look again to ERISA for another factor at play here. The movement to individual accounts started by ERISA empowers the individual to control his or her financial future. These accounts have also brought many new players into the market as financial advisors – mutual fund families, brokerage houses, and insurance companies. Many of these institutions possess trust powers through subsidiaries and aggressively offer trust services to their clients, positioning themselves as facilitators and helpers for the client’s trust beneficiaries just as they were for the client during life.

The result of these changes is a widening gulf between corporate fiduciaries and the people who need them. More and more often the client wishes to choose individual trustees even though they are less qualified than available corporate fiduciaries. Again it is time for the Third Paradigm trust, which would be drafted with a view toward bridging the gulf between grantor and professional trustee, and increasing the willingness to use the best qualified entity as fiduciary.

4. Dealing with the Challenge of Total Return Investing

How are we to handle the challenge of facilitating the use of total return investing? How do we address the problems created by the artificial distinction between income and principal? Remember that the adoption of the Uniform Prudent Investor Act may authorize total return investing, but the trust language must also permit it and facilitate it. There appear to be three options – rely upon the new principal and income act, give broad powers to the trustee to make things right, or eliminate the distinction entirely by using a non-charitable, private unitrust. Each is worthy of a closer look.

1. Rely Upon the New Uniform Principal and Income Act

There are three sources which seek to bring total return investing into the mainstream of fiduciary law. We have already mentioned two of them – the Restatement (Third) of Trusts and the Uniform Prudent Investor Act. The third is the Uniform Principal and Income Act of 1997 (hereinafter referred to as the "1997 Act"). One purpose of this act is to modernize many of the default rules governing accounting for trust investments, such as the treatment of distributions from pass-through entities such as S corps, partnerships and limited liability companies.

Another purpose of the 1997 Act, the key purpose for our discussion, is stated as follows:

The other purpose is to provide a means for implementing the transition to an investment regime based on principles embodied in the Uniform Prudent Investor Act, especially the principle of investing for total return rather than a certain level of "income" as traditionally perceived in terms of interest, dividends and rents.

The 1997 Act seeks to implement total return investing through Section 104. This section authorizes a trustee to make adjustments between principal and income when necessary if the income portion of the trust’s total return is too small or too large because of the investment decisions made by the trustee under the Prudent Investor Rule.

One response to the challenge of total return investing would be to continue to draft trusts using tried and true First Paradigm language. We could then rely upon passage of the 1997 Act to furnish the trustee with the tools to do the job we expect. This course of action would probably be a mistake for a number of reasons:

• The 1997 Act has been adopted in less than eight states, and sometimes without Section 104.

• Several states have considered, and rejected, the 1997 Act, primarily because of opposition from professional trustees to Section 104.

• The New York study commission considering the Act has proposed to define the word "income" in the phrase "pay the income of the trust to A" to mean a 4% unitrust interest.

• There can be no assurance what action, if any, your state will take on this legislation.

We must deal head-on with the challenge of total return investing. Our duty to our clients indicates that we must act, not wait for our legislature to take care of the problem in a satisfactory manner.

1. Give Broad Discretionary Powers to Trustee

Another solution to the artificial distinction between income and principal is to give the trustee a broad discretionary power to ignore that distinction. The trustee could be given total discretion to distribute income and principal among the beneficiaries of the trust. For example:

The trustee, in its sole discretion, may pay amounts of income or principal to or for the benefit of A during A’s lifetime. Upon A’s death, the trust shall be paid over to B.

The problem with this solution to our principal/income problem lies in two areas:

• It gives the trustee broad discretion to act, but this makes the client very nervous.

• It gives the beneficiary no framework upon which to build expectations.

As we have stated before, clients today tend to view professional trustees as big, impersonal, amorphous blobs, not as talented individuals. Many clients cringe at the suggestion that the trustee be given total discretion over what their beneficiaries may get from the trust. True, the grantor may give guidelines to the fiduciary, whether in the trust instrument or in a letter accompanying it. But the grantor still may question who will interpret those guidelines in the years to come. There may be situations such as spendthrift trusts where discretion is the only answer, but it is not the answer preferred by most clients and their families.

The creator of the trust may use it for tax reasons; he may use it to control the property and its ultimate disposition; he may use it to protect the beneficiary from his own judgmental errors. In each of these situations, however, the grantor does provide for one or more beneficiaries about whom he cares and whose happiness is at least one of his motivating forces. Rarely does the grantor use a trust with the intention to make the beneficiaries unhappy. The expectations of the trust beneficiaries are important in determining if they are happy with the trust, and those expectations arise from the trust instrument. (And those expectations are also heavily influenced by the concept of total return investing, as we have already seen.) The totally discretionary trust gives the beneficiaries no framework for their expectations. The beneficiaries are placed in what they perceive to be an adversarial relationship with the trustee. Make the trustee mad, or even fail to make the trustee happy, and you may receive nothing.

1. Use a Unitrust Payment Formula

Another alternative for implementing total return investing in the Third Paradigm trust, the best alternative I believe for the majority of trusts, is to use a unitrust payment formula. Such a formula would require the trustee to pay to the beneficiary a specific percentage of the value of the trust assets each year, just as in the case of a charitable remainder trust under §664 or a grantor retained unitrust under §2702. The benefits of such a trust would be as follows:

• The expectations of the beneficiaries are clearly defined in the document.

• The trustee is relieved of concerns about the duty of impartiality, for both the current beneficiary and the remainder beneficiary share in the funds’ growth (or loss). The beneficiaries are partners.

• The trust can invest in any asset without regard to what it produces as "income" or "capital appreciation."

Like any broad statement, there are exceptions to the rule; the unitrust form of payout cannot be used blindly in every case. Four types of trust in particular are not well suited for the unitrust payout formula:

• A credit shelter trust where the surviving spouse is the beneficiary and is alive. Any mandatory distributions would remove assets from the shelter of the trust and subject any unspent portion to tax in the estate of the surviving spouse.

• A generation skipping trust which provides for distributions to a non-skip person for life and then to the skip beneficiaries. The mandatory unitrust formula would be contrary to good tax planning, for any unspent portion of the mandatory distributions to non-skip persons would be taxed earlier than necessary.

• A spendthrift trust designed to protect assets from the claims of creditors. A mandatory unitrust interest would be reachable by creditors under the laws of most states.

• A trust funded largely with non-financial assets such as real property or with non-liquid assets such as interests in a closely-held business. The unitrust formula would produce an inappropriate result of requiring payments in kind or requiring forced sale of some of the assets.

The unitrust provision should be strongly considered, however, in the vast number of ordinary trusts used in estate planning. This would include irrevocable life insurance trusts after the maturity of the policy, credit shelter trusts where the spouse is not a beneficiary or where the trust continues after the death of the surviving spouse, generation skipping and dynasty trusts where no non-skip person is a beneficiary, and the typical protection trust that holds assets for X until he or she attains the desired age.

Some commentators have pointed out that the terms of the unitrust require a payout from the trust even if the beneficiary does not need the money. Two points should be made. First, because of the compressed income tax brackets applicable to trusts, all trust income over $8,900 is taxed at 39%. Distribution of these amounts to the beneficiaries generally results in a reduced income tax burden for the family group. As for the estate taxes, any funds not needed by the recipient could be given to objects of bounty using annual exclusion gifts.

Secondly, very few people who are the current income beneficiaries of a trust will be content to allow all income and gain to be accumulated in the trust, whether they "need" it or not. They want something distributed to them, and they will find the "need" for it. Given this aspect of human nature, the use of a unitrust payout formula imposes no real financial hardship on the trust or its beneficiaries.

1. The Private Unitrust – Design Consideration.

A private unitrust would closely resemble the charitable remainder unitrust with which we are all familiar. Let us look at selected provisions of such a trust and discuss the choices to be made by the drafter in consultation with the client.

1. The Distribution Amount in the Ordinary Unitrust

We must first define the unitrust amount to be paid to the beneficiary each year. Consider the following language:

The Distribution Amount shall be an amount equal to ___ percent (___%) of the fair market value of the trust as of the close of the last business day of the preceding tax year of the trust.

Perhaps the most difficult question is, what percentage should the payout be? Selection of the payout percentage should start with consideration of factors including the beneficiary’s needs, the period the trust is to continue, and the grantor’s desire for the remainder beneficiaries (e.g., should they receive anything, receive the same principal as the trust started with, receive the same principal adjusted for inflation, or receive an increased principal after adjustment for inflation).

Many studies indicate that, over an extended period, a percentage of 3 to 5% is appropriate to protect the current beneficiary while passing to the remainder beneficiaries an inflation-adjusted principal equal to the starting amount. Clients who have seen the unprecedented performance of the stock market in the last decade may at first believe these suggested rates to be absurdly low. The rates are, however, realistic when the time line is longer, and the life of most trusts is longer.

When considering the percentage to be used, remember that what goes up may also come down. In years when the market value of the trust declines, the distribution amount will also decline. Rarely, however, do the expectations of beneficiaries decline in concert with the financial markets. Accordingly, using an average value of the trust over a period of years may be used to smooth out changes in market value. Such a variation might be as follows:

The Distribution Amount shall be an amount equal to ___ percent (____%) of the average of the fair market values of the trust as of the close of the last business day of the three (3) preceding tax years of the trust.

This provision will reduce the amount of variation in the distribution amount and aid the budgeting of the beneficiary and the cash flow projections of the trustee.

Also remember that there is no requirement in the private unitrust that the Distribution Amount remain fixed throughout the term of the trust. It can be changed to accommodate different stages of the life of the beneficiary. For example, consider the following clause:

The Distribution Amount shall be an amount equal to a percentage of the fair market value of the trust as of the close of the last business day of the preceding year of the trust, the percentage being determined as follows: (i) until _____ attains the age of forty (40) years, three per cent (3%); (ii) Between the ages of forty (40) and fifty five (55) years, four per cent (4%); and (iii) after attaining age fifty five (55) years, five per cent (5%).

Such a distribution formula might be intended to minimize distributions during the prime working years of the beneficiary, then to increase the payout to assist with college expenses, finally to increase again to supplement retirement.

The unitrust payout could also be introduced into the trust after the happening of a stated event. For example, a credit shelter trust might provide for income and principal to be made available to the surviving spouse during his or her lifetime in the trustee’s discretion. After the death of the spouse, however, the trust could provide for the payment of a unitrust interest to the new beneficiary.

2. A Variation – the Percentage Unitrust.

The definition of the Distribution Amount as a dollar amount is simple and straightforward. There are, however, at least two variations which have been suggested by Jerold Horn to be considered.

1. The Mandatory Percentage Unitrust

One alternative is illustrated in the following definition of the Distribution Amount, here called the Annual Distribution and expressed as a percentage of the trust:

The Annual Distribution shall be ____ percent (____%) of the trust estate as of the close of the last business day of the preceding tax year of the trust.

This definition converts the dollar amount of the ordinary unitrust into a percentage share of the trust. There should be two advantages to this model. The trustee can make the distribution in kind, and, unless the election under §643(e) is made, there should not be a deemed sale. Of course, the recipient takes a carry-over basis in the property if the election is not made. Thus, by making or not making the election, the trustee can accelerate or defer the recognition of gain and determine who, the trust or the beneficiary, will pay the tax.

The second advantage is that the ability to distribute in kind permits the trustee to remain fully invested, without maintaining a store of cash to pay the distribution amount or without selling an investment. However, the distribution in kind will carry out DNI to the distributee under §§661 and 662, and increase his or her taxable income, possibly requiring some cash to pay the resulting income tax.

To make this concept fulfill its potential as a planning tool, the trust document should grant to the trustee the power to make non-prorata distributions in kind, either explicitly in the instrument or by the incorporation by reference of an appropriate statutory power, or both.

2. The Withdrawal Percentage Unitrust

Another variation of the percentage unitrust discussed by Horn eliminates the element of mandatory distribution and permits the beneficiary the discretion to take some, all or none of the Distribution Amount. This definition might read as follows:

If the beneficiary is living immediately before the end of a calendar year, the trustee shall pay to the beneficiary so much, if any, of the trust estate as the beneficiary shall direct in writing before the end of the year, but the portion withdrawn shall not exceed ___ percent (___%) of the trust estate.

Such a unitrust might be desirable if there is a question of whether the beneficiary will need the Distribution Amount, or if he or she would prefer to compound those amounts. Of course, the power of withdrawal granted here is a general power of appointment, but its lapse would be shielded from estate and gift tax consequences by the 5 x 5 exception of §2041(b) and §2514(e) if the percentage withdrawal is set at 5% or lower. However, all income attributable to the property subject to the power will be included in the income of the beneficiary under §678(a)(2).

It must be remembered that the 5x5 exceptions are only applicable for transfer tax purposes. The IRS has taken the position that the lapse of such a power results in the power holder being considered the grantor of the trust for income tax purposes to the extent of the property not withdrawn. Thereafter, depending upon the terms of the trust, the holder of the power may be subject to the grantor trust rules as to an ever-increasing portion of the trust corpus.

1. Qualification for the Marital Deduction

The Internal Revenue Code still talks of principal and income in many places, but particularly in §2056 dealing with the martial deduction. In §§2056(b)(5) and 2056(b)(7), dealing with the power of appointment trust and the QTIP trust respectively, the Code requires that the surviving spouse receive all of the net accounting income of the trust each year. Therefore, if the trust being drafted as a private unitrust is intended to qualify for the marital deduction, the language set forth below should be added:

If, in any year, the net income of the trust exceeds the Distribution Amount, such excess shall be paid to A at least annually.

Of course, this provision could terminate with the death of the spouse if the trust is to continue for other beneficiaries.

2. Additional Discretionary Distributions

Just as in any First/Second Paradigm trust, it may be desirable to vest in the trustee discretion to pay additional amounts to the beneficiary to meet circumstances not covered by the mandated distributions. This power of distribution can be defined as broad or as narrowly as the drafter and the client may choose. Consider the following provision:

In addition to the Distribution Amount, my Trustee may, in his sole discretion, distribute to or for the benefit of A such amounts from the income or principal of this trust as the trustee deems advisable for her health, education, maintenance and support in reasonable comfort.

This might be considered the best of both worlds. The proponents of the "give it to the trustee with total discretion" school could here have whatever their favorite discretion language is, deriving all of the admittedly real benefits for so doing. The trustee would be permitted to meet financial emergencies of the beneficiaries or to deny additional distributions if circumstances dictated that course. At the same time, much of the pressure on the trustee in terms of investment strategy and dealing with the beneficiaries is eliminated by the mandatory floor for distributions set by the unitrust formula.

3. Source of Distributions

The drafter should also give guidance as to the income tax ramifications of the amounts distributed. Consider this language taken from charitable remainder unitrust forms:

The amounts distributed from the trust shall betreated under the rules set forth in Regulations §1.664-1(d)(1) as that Regulation exists upon the date hereof. Any capital gains actually distributed as part of the Distribution Amount or as a discretionary distribution shall be credited to the recipient of the distribution on the books of the trust.

This paragraph makes clear the source of the payments made in satisfaction of the Distribution Amount. While there is no law explicitly making them applicable to private unitrusts, the IRS has made it clear that the trust instrument may set forth ordering rules.

1. The Other Challenge – Providing A Trustee

Every trust must have a trustee. That trustee must act as defined in applicable trust law, but must also deal with the factors unique to each trust. A corporate trustee is best suited to be trustee of the Third Paradigm trust. But we have already spoken of the growing reluctance (even outright refusal) of clients to use them. The Third Paradigm trust must therefore be designed to bridge the gap between these two poles - the expertise and impartiality of the professional trustee and the reluctance of the client to use them.

A. What Is the Problem

Why are clients sometimes reluctant to use a corporate fiduciary? This reluctance does not arise primarily from a concern about fees. Fees of the corporate fiduciary are usually no more that those of other investment advisers, and frequently less. In a very real sense, when you use a corporate trustee, you pay the same fee for investment services that you would pay without such a trustee, and you get the fiduciary services thrown in for good measure.

But the reluctance does arise from anecdotal evidence of dissatisfaction with other trusts and from a perception of the qualities of the modern bank trustee. They are perceived as big, governed by committee, slow to act, with frequent change in the contact person leading to the "trust officer of the month" syndrome, and an overriding aversion to risk. I fear that some of these perceptions may be well-founded!

The answer to bridging the gap may be to consider carefully the powers that can be retained by the grantor, or at least given to a third party trusted by the grantor. These powers make the grantor feel less "powerless" after the trust is established and less "at the mercy of" a big institution. Thus he or she may be more inclined to use a corporate trustee to take advantage of the areas like investment management, tax compliance, and record keeping where the bank has acknowledged expertise.

1. A Statement of Purpose for the Trust.

The Third Paradigm trust should start with a statement of the purpose of the trust, a plain language explanation of why the grantor is creating the trust. While this may be challenging to the drafter of the document, it would accomplish three things:

It would serve as a basis for the expectations of the beneficiaries with regard to distributions from the trust. We have previously mentioned the role of these expectations in determining the satisfaction of the beneficiaries.

It would give guidance to the trustee for the exercise of any discretionary powers contained in the instrument. The primary role of the trustee is to serve as a surrogate for the absent grantor, and a statement of purpose with enhance the trustee’s ability to fill this role.

It would hopefully protect the trust and/or the trustee from modification of the trust in the future. It has long been the law that the grantor and all beneficiaries could ban together to modify an otherwise irrevocable trust. However some states have gone beyond this to provide for less stringent requirements for such modification. A statement of purpose in the document should carry great weight in any proceeding seeking to change the purpose of the trust.

C. The Power to Remove and To Replace the Trustee

There is one broad power which can, and should, be retained by the trust grantor without adverse tax consequences under the estate tax, gift tax, or income tax statutes. That is the power to remove and replace the trustee. Since the trustee carries out the terms of the trust, how it operates and the attitude it employs in its operations are the most important things in determining the trust’s success. The grantor should have all permitted power to control the identity of the trustee who in turn controls the trust which he or she has established.

If the grantor has an unrestricted right to remove a trustee and to appoint himself trustee in that place, the powers of the trustee will be attributed to the grantor. Therefore if the trust gives the trustee any power that would, if retained by the grantor, trigger inclusion of the assets in his or her estate under §§2036 or 2038, the trust corpus will be included in the grantor’s estate.

The present state of the law in set forth in Rev. Rul. 95-58, 1995-2 C.B. 191. That ruling provides that trust assets will not be included in the grantor’s estate under §§2036 or 2038 if the power permits removal of the trustee and replacement only with an individual or corporate trustee who is "not related or subordinate to the [grantor] (within the meaning of Section 672(c))." Under §672(c), a related or subordinate party includes the grantor’s spouse, mother and father, lineal descendants and employees. Also included are any corporation (or employee of that corporation) in which the grantor is an executive or significant voting shareholder.

Given the breadth of this ruling, a trustee removal power should be included in virtually every trust. This power may be retained by the grantor during his or her lifetime, and upon death, it can pass to one or more persons of the grantor’s choosing. In drafting the power, the following questions should be considered:

• Who among the possible beneficiaries has the power to remove and replace? Perhaps one has the power to remove and another to replace.

• If more than one have the power, how is the decision made? By majority vote? Or does one individual have a tie-breaking vote?

• Should the exercise of the power be limited to a finite number of times?

• How is the class of permissible replacement trustees defined? Is it limited to corporate trustees or could individuals be appointed?

• As to corporate candidates, are there are qualifying conditions such as assets under management or jurisdiction of governing law?

• As to individual candidates, are there any qualifying conditions other than that they not be related or subordinate under the meaning of §672(c)?

There is no power more important than this one in the drafting of the Third Paradigm trust!

1. The Trust Protector (or Special Trustee or Trust Advisor)

There are numerous other powers that the typical grantor would like to have - powers such as the right to add or delete beneficiaries or the right to alter the dispositive terms of the trust. These powers, however, create severe estate and income tax problems for the grantor. They should be given to someone else who could then exercise them in accordance with the perceived intentions of the grantor.

For many years, legal practitioners in English common law jurisdictions have made regular use of a special trustee, often called the trust protector, to add flexibility to trusts. The role of the trust protector is to modify the terms of the trust when necessary to carry out the grantor’s intent.

The powers of the trust protector must be defined in the trust document. The power can be simple, such as a right to remove and replace trustees, or it can be more complex, such as a power to alter the beneficial enjoyment of the trust among a class of people. There is no limit to the varieties of these powers; each trust drafter can fashion them in many ways to serve the needs of the clients and of the trust beneficiaries.

Powers that may be granted to a trust protector and that might be desired by the grantor would include:

• The power to amend the dispositive terms of the trust, such as adding or removing beneficiaries or changing ages of distribution.

• The power to amend the administrative terms of the trust, such as removing and replacing the trustee.

• The power to appoint his or her own successor as trust protector.

• The power to exercise discretion as to the recipient, time, nature and amount of distributions from the trust itself.

• The power to change the legal situs of the trust.

1. Who can be a Trust Protector

The powers given to the trust protector can be broad and sweeping, so we must be concerned with the tax impact of them. The first question would be, who can serve as trust protector? Clearly, the grantor should not. Many of the powers given to a trust protector, other than the power to change trustees, would surely cause the trust corpus to be included in the grantor’s estate under §§2036(a)(2) or 2038(a)(1).

Moreover, the grantor should not have a power to remove and replace the trust protector. Under the rationale of Rev. Rul. 95-58, the Service can attribute the powers of a trustee to the grantor who holds a power of removal and replacement, if the successor can be a related or subordinate party. Given the broad powers of the trust protector, it is probably best to avoid any temptation to the Service to extend the rationale of that ruling to these facts.

For transfer tax purposes, almost anyone other than the Grantor, and specifically including a spouse or child, can be named as trust protector. Indeed, a good argument could be made that the ideal trust protector is the spouse of the grantor. Who else would have such an intimate knowledge of the grantor’s desires and wishes in establishing the trust? And who else who know the beneficiaries, their strengths and their weaknesses so well? Regulations. §20.2036-1(b)(3) provides that its provisions do not apply to powers held by a person other than the transferor. So do the Regulations under §2038. Case law also supports the statement that the spousal powers will not cause estate tax questions. In Kneeland v. Com’r, the decedent had established an irrevocable trust for children and grandchildren. The grantor’s spouse was given a power to revoke the trust and return the trust corpus to the grantor. The court held that the trust was not included in the grantor’s estate because the spouse’s power did not require that result. Of course, in none of these cases or Regulations was there any agreement or understanding between the grantor and the spouse as to the exercise of the power, and there should be none!

It is important to recognize that some of the powers given to the trust protector may be sufficient to constitute a general power of appointment in the holder. Thus if the power given to the spouse permitted him to vest title to the property in himself, his creditors, his estate or the creditors of his estate, that would be a general power and the trust corpus would be included in his estate. However, if that power is limited by an ascertainable standard, it will not be a general power, and no inclusion will result.

If there is no spouse available, or if the grantor desires the trust protector to have powers that could cause estate tax inclusion, the trust protector should be unrelated individual who is not subordinate within the meaning of §672(c). Lineal descendants of the grantor are not a good choice since they are normally beneficiaries of the trust itself. As beneficiaries, their exercise of powers may be challenged as biased toward themselves and the powers may constitute general powers of appointment with estate and gift tax consequences. All in all, not a good choice. This points us to an unrelated individual or perhaps a corporate entity.

For income tax purposes, the rules are different from those we have discussed above.. Section 672(e) provides that the grantor of a trust will be treated as holding any power held by his or her spouse. Moreover, any power held by a "related or subordinate party" will be deemed held by the grantor. See §672(c). Most of the powers granted to a trust protector, other than purely administrative powers, would or could be sufficient to make the grantor owner of the income of the trust under §§673 to 677 of the Code.

Accordingly, unless the trust protector is not the spouse or a "related or subordinate party," the trust likely will be a grantor trust, causing all items of income and gain to be taxed to the grantor. But consider (i) the trust protector can have the power to make money available to the grantor with which to pay the tax, and (ii) the payment of the tax liability would further reduce the grantor’s estate without the making of a gift.

All things being said, a general rule is that the trust protector should not be the grantor or a beneficiary under the trust document. Otherwise, the field is wide open!

F. A Word of Caution

In many instances, the Internal Revenue Code – the ruler of the Paradigm Two trust – provides exact and detailed requirements which must be met if a trust is to perform its intended purpose. If you desire to use a trust protector in such a trust, you must be sure that the powers granted to the protector do not include the power to change any of the statutory requirements for the trust. Some examples may be helpful:

• In trusts intended to qualify for the marital deduction under §2056 (except for an estate trust), the Internal Revenue Code requires that all income be paid to the spouse each year and that no other person be a permissible beneficiary of the trust during the lifetime of the spouse. Any trust protector should, therefore, have no power to violate either of these requirements.

• Similar requirements are imposed on a trust which intends to be a Qualified Subchapter S Trust (QSST) under §1361(d). Again the protector’s powers should be limited to prevent any possible violation of the Code requirements.

Other types of trusts that have specific tax requirements would include, be not be limited to, the following:

• A §2503(c) trust for minors;

• A direct skip trust qualifying for the annual exclusion under §2642(c);

• A charitable remainder trust under §664;

• A grantor retained income trust under §2702;

• An Electing Small Business Trust (ESBT) under §1361(e); or

• Any trust with Crummey withdrawal powers, at least with respect to any existing power.

In structuring a trust protector provision in such an environment, a two-pronged approach is probably desirable. First, the specific powers granted to the trust protector should be carefully examined to see which should be eliminated and which modified. Then a savings clause could be added to the provisions similar to that set forth in Paragraph E of Exhibit B.

VII. Summary

For too long we lawyers have created trusts without devoting sufficient time and energy to the question of how they will operate in the future. Today, however, external forces in the nature of wealth, the nature of investment management, the nature of the taxation of wealth, and the nature of the trust manager force us to take another look at our tried-and-true drafting habits.

The conclusion reached from this review is that we should change what we do. We should consider alternative forms of trust distribution formula, especially the private unitrust. We must also consider the changing nature of the institutional trustee, and draft trusts better adapted to the needs of the grantor in this environment. Provisions such as the trust protector provisions permit us to blend the best skills of the institutional trustee with those of an individual familiar with the grantor’s wishes to achieve the optimum result for our clients.

EXHIBIT A

LAW IN NORTH CAROLINA:

PRUDENT INVESTMENTS AND TOTAL RETURN INVESTING

In 1999, North Carolina joined the majority of states in this country which have adopted the Uniform Prudent Investor Act. This Act, codified as Article 15 of chapter 36A of the North Carolina General Statutes, basically provides that the trustee should invest the trust assets "as a prudent investor would, by considering the purposes, terms, distribution requirements, and other circumstances of the trust." N.C.G.S. §36A-162(a). From this rule and the others stated in this Article, we see the following principles which govern trustee behavior:

• The trustee should focus on total return; the distinction between income and principal is lessened in importance. (§36A-162(c)5.)

• The trustee should focus on the entire portfolio, not individual assets within it. (§36A-162(b).)

• The trustee may invest in any kind of property so long as it is prudent to do so. (§36A-162(e).)

• The trustee must review the assets of the trust upon assuming office, and decide to retain or dispose of them to bring the portfolio into compliance with this rule. (§36A-164.)

• The trustee may delegate authority for investment decisions if it is prudent to do so and if the agent is selected with reasonable care. (§36A-169(a).)

• The trustee has a duty to diversify the portfolio unless, under the prevailing circumstances, it is prudent not to do so. (§36A-163.)

This statute is effective January 1, 2000, and applies to express trusts existing on and created after the effective date, unless the terms of the trust are to the contrary. (§36A-171.)

EXHIBIT B

LAW IN NORTH CAROLINA

MODIFICATION OR TERMINATION OF IRREVOCABLE TRUSTS

In 1999, North Carolina adopted a new statute offering an opportunity to modify or terminate irrevocable trusts in certain circumstances. The statute, codified as Article 11A of Chapter 36A of the General Statutes, provides as follows:

• If the grantor of the trust is the sole beneficiary, he can compel modification or termination of the trust at any time. (§36A-125.2)

• If the grantor and all beneficiaries agree, they can compel the modification or termination of the trust without approval of a court. (§36A-125.3)

• If one of the beneficiaries of the trust does not consent, the others may seek modification or partial termination from the court. The court can approve the relief requested if it finds no harm or impairment to the beneficiaries who do not consent. (§36A-125.3(b))

• If all beneficiaries of the trust consent, they may seek modification or termination of the trust in a court proceeding, presumably even if the grantor is alive and does not consent. (§36A-125.4(a))

• If the actions desired by the beneficiaries would effect a material purpose of the trust, the trust cannot be modified or terminated unless the court determines "that the reason for modifying or terminating the trust under the circumstances substantially outweighs the interest in accomplishing a material purpose of the trust." (§36A-125.4(b))

• In its considerations, the court may modify or terminate the trust if (i) the purpose of the trust has been fulfilled, or become illegal, or become impossible of fulfillment; or (ii) owing to circumstances unanticipated by or unknown to the grantor, the continuation of the trust would defeat or impair the accomplishment of the purposes of the trust. (§36A-125.7(a))

EXHIBIT C

PROVISION FOR A TRUST PROTECTOR

[Adapted from a form by Hodgman in his article cited in the Bibliography]

1. I appoint _________________ as Trust Protector hereunder. If ____________ fails or ceases to act as Trust Protector, he may, but is not required to, appoint any one or more successor Trust Protector as provided in the following paragraph. No trust created under this instrument is required to have a Trust Protector acting with respect to that trust.
2. The Trust Protector acting from time to time, if any, may appoint any one or more individuals (other than me, my spouse, or a descendant of mine) as successor Trust Protector. Any appointment of a successor Trust Protector shall be in writing, may be made to become effective at any time or upon any event, and may be single or successive, all as specified in the instrument of appointment. The Trust Protector may revoke any such appointment before it is accepted by the appointee, and may specify in the instrument of appointment whether it may be revoked by a subsequent Trust Protector. In the event that two or more instruments of appointment or revocation by the same Trust Protector exist and are inconsistent, the latest by date shall control.
3. Any Trust Protector may resign by giving prior written notice to the trustee. All trusts created under this instrument need not have or continue to have the same Trust Protector. The provisions of this instrument that relate to the Trust Protector shall be separately applicable to each trust held hereunder.
4. The Trust Protector may, with respect to any trust as to which the Trust Protector is acting, modify or amend:

1. The trust administrative provisions of Article ___ relating to the identity, qualification, succession, removal and appointment of the trustee;
2. The financial powers of the trustee set forth in Article _____;
3. The provisions of Article _____ relating to the identity of the contingent beneficiary of the trust property;
4. The withdrawal rights granted under Article ____ of this instrument (except a withdrawal right already in existence at the time the Trust Protector seeks to exercise the power conferred under this paragraph); and
5. The terms of any trust created hereunder with respect to (i) the purposes for which the trustee may distribute trust income and principal, and the circumstances and factors the trustee may take into account in making such distributions, (ii) the time for the distribution or withdrawal of any portion of the principal or income of the trust, (iii) the termination date of the trust, either by extending or shortening the termination date (but not beyond any applicable perpetuities period), and (iv) the identity of the permissible appointees under the testamentary power of appointment granted to the beneficiary for whom the trust is named.

1. The provisions of this Article shall not apply to any trust created hereunder which is intended to qualify for the marital deduction under the provisions of §2056 of the Internal Revenue Code or which is intended to qualify for the charitable deduction under any provision of the Code.
2. The rights and powers conferred on the Trust Protector under this instrument, including, without limitation, the power to remove trustees, and all rights and powers granted the Trust Protector under Paragraph D of this Article, shall be exercisable only in a fiduciary capacity.
3. Notwithstanding any other provision of this instrument, the Trust Protector shall not participate in the exercise of a power or discretion conferred under this instrument that would cause the Trust Protector to possess a general power of appointment within the meaning of Sections 2041 and 2514 of the Code.
4. The Trust Protector acting from time to time, if any, on his or her own behalf and on behalf of all successor Trust Protectors, may at any time irrevocable release, renounce, suspend, cut down or modify to a lesser extent any or all powers and discretions conferred under this instrument by a written instrument delivered to the trustee.

BIBLIOGRAPHY

The following Bibliography lists the articles most important to an understanding of the thoughts presented here. I suggest that these articles be read to give you a flavor of the excellent research which has been done in this area, for all of them have contributed to my understanding of these issues. (Errors, however, are all mine.) From time to time, specific reference is made in the notes to articles cited here when some particular fact or language is extracted. That citation is to the name in brackets following each article.

Darin N. Digby, "What Powers Can a Donor Retain Over Transferred Property?", 24 ESTATE PLANNING 318 (Aug./Sept. 1997). [Digby]

Joel C. Dobris, "New Forms of Private Trusts for the Twenty-First Century – Income and Principal," 31 REAL PROP. PROB. & TRUST LAW J. 2 (1996). [Dobris I]

Joel C. Dobris, "Changes in the Role and the Form of the Trust at the New Millennium, or, We Don’t Have to Think of England Anymore," 62 Albany L. Rev. 543 (1998). [Dobris II]

Mark B. Edwards, "Trusts for the Third Century: The Third Paradigm," 18 THE WILL AND THE WAY, No. 1 (November 1998). [Edwards]

James P. Garland, "The Problem with Unitrusts," THE JOURNAL OF PRIVATE PORTFOLIO MANAGEMENT, No. 4 (Spring 1999)[Garland]

David R. Hodgman, "Drafting Flexible Irrevocable Trusts – Whom Can You Trust?", 23 ESTATE PLANNING 221 (June 1996). [Hodgman]

Graham D. Holding, Jr. and Christy Eve Reid, "The Private Unitrust vs. The Discretionary Trust as a Paradigm for the New Century," 18 THE WILL AND THE WAY, No. 2 (Feb. 1999). [Holding/Reid]

William L. Hoisington, "Modern Trust Design: New Paradigms for the 21^st Century," 1997 U. MIAMI INST. ON ESTATE PLANNING 6-1 (1997). [Hoisington]

Jerold I. Horn, "Prudent Investor Rule, Modern Portfolio Theory, and Private Trusts: Drafting and Administration Including the "Give-Me-Five" Unitrust," 33 REAL PROP. PROB. & TRUST LAW J. 1 (Spring 1998). [Horn]

David W. Keister & William J. McCarthy, Jr., "1997 Principal and Income Act Reflects Modern Trust Investing," 26 ESTATE PLANNING 99 (March/April 1999). [Keister/McCarthy]

Jonathan R. Macey, AN INTRODUCTION TO MODERN FINANCIAL THEORY (2^nd ed. 1998). [Macey]

Robert J. Rosepink, "The Total Return Trust – Where and How to Tax Capital Gain, 137 TRUSTS & ESTATES 12 (Oct. 1998). [Rosepink]

Robert B. Wolf, "Defeating the Duty to Disappoint Equally – The Total Return Trust," 32 REAL PROP. PROB. & TRUST LAW J. 45 (Spring 1997). [Wolf I]

Robert B. Wolf, "Total Return Trusts - Can Your Clients Afford Anything Less?" 33 REAL PROP. PROB. & TRUST LAW J. 133 (Spring 1998). [Wolf II]

Robert B. Wolf & Stephen R. Leimberg, "Total Return Unitrusts: The (TRU) Shape of Things to Come," 10 RIA EST. PLANNER’S ALERT 15 (Dec. 1998). [Wolf/Leimburg]