Heckerling Institute Report
Report #8 - Total Return Trusts
Back
to Heckerling Table of Contents
The following is Report #8 from our on-site
reporters regarding some of the highlights from the events and presentations that are
taking place at the 34th Annual Philip E. Heckerling Institute on Estate Planning that is
being held January 10-14, 1999 at the Fontainebleau Hilton in Miami Beach, Florida.
First, an eratta re Report #7. Lou Mezzullo's presentation on "Entities"
took place first thing Wednesday morning, not on Tuesday as was mistakenly reported in
Report #7. We apologize in advance for any confusion this may have caused anyone,
and especially Lou.
This Report covers Wednesday afternoon's Special Session I-C entitled "The
Total Return Trust [TRU] Revolution: An Introduction" that was co-presented
Wednesday afternoon by Robert Wolf, Stephan Leimberg and Susan Porter,
and Special Session II-A entitled "CASE STUDY - Total Return Trusts:
Techniques and Applications" that was presented by Robert Wolf and Bruce
Guiot.
The portion of this report on Special Session I-C was filed by none other than on-site
reporter, Stephen Leimberg, who was one of the co-presenters of that Special Session.
The portion of this report on Sepcial Session II-A was filed by none other than
on-site reporter, Robert Wolf, who was one of the co-presenters of that Special Session.
Nothing like getting it straight from the horses mouth we always say, especially
when it comes to a multi-facited topic such as this one. This is a long post too,
but well worth reading. If for any reason this report is too big for your e-mail
system to handle properly, you can always view it in full text in the searchable Web-based
archive for this list, which is located at URL <http://home.ease.lsoft.com/archives/aba-ptl.html>.
Summary - Special Session I-C: Wolf and Leimberg began this Special Session
by explaining the modern portfolio theory, defining total return investing, and comparing
the TRU (Total Return Unitrust) with the traditional "income payout" ,
"discretionary", and indexed annuity trust design. They explained that,
although the TRU was an exceptionally important planning tool that should be considered in
any trust situation, it was not a "trust for all seasons" but rather one of many
tools to be applied where and to the extent appropriate. They emphasized that one of
the most important elements of planning in the decades to follow was the importance of
drafting to meet client-monetary as well as tax savings goals - and appraising the client
of realistic expectations after considering computer modeling which takes into
consideration trust expenses, turnover, inflation, and taxes.
Details - Special Session I-C: Wolf and Leimberg made the following points
during their presentation:
TRUs can solve problems that can't be solved by the traditional trust model.
• Future planning and drafting will be based more on our client's needs and
desires.
• Total return planning will reinforce our client's financial and investment
planning.
• Total return planning and computer modeling will highlight insufficiency in
capital to accomplish stated objectives.
• Total return planning will be driven by human needs--not taxes - yet
it will help us plan better for taxes too.
• The usual approach to drafting trusts - telling the trustee to "Hold
the principal - and pay the income" in what we call the "income rule trust"
-- is not just the wrong thing to do: From a long-term financial
viewpoint, drafting the traditional income rule trust may be as harmful - or even more
harmful - than making a drafting error that causes trust principal to be included in a
beneficiary's estate.
• Articles in Standard and Poor's Outlook, Bloomberg Personal Finance,
and FORBES MAGAZINE have made the public aware - not only that there is a
choice - but that the choice can make a lot of
difference!
• Although our clients come to us to create trusts set to last for many years
or many generations, those well-intentioned trusts - by definition and design
cause conflict between those generations
• This potential for family conflict is inherent in the design itself.
It's because trusts create separate categories of beneficiaries, current and future - and
the interests of the two are instantly and inevitably in competition. This also
places the trustee in a very unenviable position: If the investment portfolio is
invested for income, the remainder beneficiaries are unhappy. If the portfolio is
tilted toward growth, the income beneficiary is upset. Quite often, a grantor's spouse
ends up on opposite sides from the grantor's children or grandchildren, the exact opposite
result from the family harmony the grantor intended, hoped for, and assumed.
• In an attempt to be fair and impartial, the trustee will typically purchase
a near equal proportion of income (typically bonds) and growth (typically equities) .
This produces neither a satisfactory level of
income (since so much is invested in equities with very low dividend yields) nor growth
(since so much is invested in bonds with no growth potential). Too often, neither
party's satisfaction will be achieved
and the portfolio will - as a whole - under-perform. This makes both sets of parties
unhappy with the trustee's performance - and the ultimate goal of the grantor - financial
security for both parties - is
thwarted.
• What's not readily apparent without computer modeling is the incredible
degree to which the classic trust model dramatically reduces a trust's potential
investment returns.
• Assuming the S&P 500 dividend yield remains at about 1.2% , on a million
dollars of stocks, a trust invested totally in equities yields (gross) $12,000 a year!
After trustee's fees are charged to income, and
taxes are paid , that $12,000 shrinks to around $8,000 a year.
• Investing in fixed income solves the income problem - but creates another
since Treasury Notes are yielding slightly more than 5.5%. So each
"unit" of $1,000,000 generates only $55,000 a year - before
trustee's fees and taxes. And that's investing every nickel - the entire trust
corpus - in fixed income investments - resulting in zero growth.
• So the trustee has a Hobson's choice -- invest for growth, and there's
almost no income or invest for income and there will be no growth.
• Traditional trust documents don't even talk about this choice. Our trusts
don't tell the trustee what the client wants - because we attorneys have never asked!
Our trusts just say, "pay all the income".
• Trustees are required to be impartial as between the life beneficiary and
the remainder beneficiary - which is why most trustees don't want to invest less than half
of a portfolio in equity securities.
• If the trustee makes a 50-50 mix, half S&P 500 - at about 1.2%
yield - and half 5 Year Treasuries - at about 5.5% return - averages out -
after trustee's fees - to approximately 3.3% current return -
and minimal growth.
• After taxes, expenses, and inflation, over the long Ibbotson period from
1926 to the present, the only category of investment that had any return at all is stocks;
bonds have no net real return--and that has
enormous implications upon what we can pay out and how we need to be drafting our
trusts!
• Yet, because of the low yield of stocks, there would not be sufficient
income for the income beneficiary. So a portfolio solely based on stocks will not
accomplish our dual objectives.
• One answer to this problem is total return investing, the achievement of the
highest overall rate of return, including both income and appreciation - consistent with
the risk tolerance of the investor.
• The TRU (Total Return Unitrust), like its charitable counterpart, is an
expanding (or contracting ) pie. The current beneficiary is given a fixed
percentage of the pie as revalued year to year. So if the pie grows bigger (or
smaller), so does the slice. For instance, if we assigned a 4% annual interest in
the pie to the current beneficiary and the pie was $1,000,000, she'd get 4% of $1,000,000,
$40,000.
• The purpose of TRU is to allow a trustee to invest for total return,
eliminate the conflicts of interest between the life beneficiary and the remainder
beneficiary, a trust which will work fairly and according to understandable and dependable
rules all the parties are aware of - no matter whether the market goes up, down, or how
fast. Finally, a TRU allows us to make the distributions to our current beneficiary
"smooth" (non-volatile) without unduly risking the trust's principal.
• Section 104 of the Revised Income and Principal Act was designed to
alleviate problems experienced by traditional trusts. It gives trustees power to
adjust returns between principal and income. And it grants that power without any
specific limitations - other than the duty of impartiality -- the duty of fairness.
This broad ability to make adjustments between the interests of the current and future
beneficiaries gives the trustee a way to increase or decrease what the income beneficiary
gets. Although many trustees will be glad for the power to make such adjustments,
many others may not want that responsibility (curse?). Nor will some beneficiaries want
such a power in someone else's hands. Instead, they (and the trust's trustee and
investment advisors) may want - a concrete rule.
• Modern Trustees must have the ability to invest for total return regardless
of what is defined as income and what is principal. But in many situations, trustees may
need and want - a distribution method, a rule--not just a power. For many
reasons, and in many situations all the parties want and need to know what the rule - or
rules are - in advance.
• There are, of course, alternatives. One of our best alternatives to
the TRU is to draft more fully discretionary trusts. Give the independent corporate
or other trustee the right to distribute or reinvest income or principal with complete
freedom, or provide discretion subject to ascertainable standards. Where the family
situation permits it, particularly where the family gets along well and
is relatively affluent, a fully discretionary trust can work very well. Sometimes, the
discretionary trust will be the very best choice!
• But absolute discretion placed in the hands of a trustee will not work
everywhere. For instance, if the grantor's spouse doesn't want to have to ask the
trustee for money, a fully discretionary trust just will not meet the client's objectives
- even with a built-in right to change trustees. And when things get tough (such as
when there is animosity between the income and the remainder beneficiary), the trustee's
job can get very difficult. The very discretion that is such an asset in some situations
can be a curse in others! Because a fully discretionary trust by definition
provides no real guidelines, each year a trustee must still decide(and neither income nor
remainder beneficiaries can be sure how that decision will be made or what it will be):
" How much should I, or How much can I - distribute?" The grantor
must be willing to give to the trustee the full control, even over the overall division of
the economic pie. This loss of control may not please the very people who the trust
was created to protect and unify.
• A second alternative to the TRU is to use an index pay out trust. The
way this works is to provide in the instrument for the initial pay-out. The trust
provides that the pay-out will be adjusted each year for inflation. In some
situations (e.g. a relatively short term trust) this can be a very appropriate choice. But
a payout linked to inflation can also be risky because the trust can be quickly exhausted
and high payouts occur by definition and design at the worst possible time.
• In the TRU the payout to the current beneficiary each year is based not on
accounting income but on a percentage of the average trust values in the 3 previous years.
This back averaging ("smoothing") avoids the potential for volatile
swings in income. For example, the payout, the distribution rate - say 4% - is
multiplied by an average of the year end values for the last three years. By using
this simple three year smoothing rule, it is possible to decrease distribution volatility
by almost 50%.
• The TRU provides an identity of interest among the three key parties to the
trust, the current beneficiary, the remaindermen, and the trustee. When one profits,
all profit just as when one suffers, all suffer together, because the distribution, the
trustee's fees and the principal value all depend on the total return from the trust--they
are all true partners in the trust venture.
• Because there is a formula -stated in advance - a rule that everyone can
understand - and easily compute - the expectations of the parties are both realistic and
likely to be met. Everyone - the trustee - the current - and the future
beneficiaries - all know how the distribution will be measured. For example, a 4%
distribution on a $1,000,000 trust will be - $40,000. There's nothing to argue about.
There's no guessing, no mystery, nothing to be disappointed about, and no decision-maker
to be angry with.
• A TRU can qualify for the marital deduction if the document specifically
requires that all trust income be distributed annually or more frequently. This is
unlikely to be a problem - since the actual unitrust pay-out amount will almost always be
in excess of the accounting income of the trust.
• An "ordering provision" requires the trust to pay out income. Then
to the extent the income isn't sufficient, the trustee is to allocate short term capital
gain to income. To the extent that's not enough, long term capital gain is forced
out. To make up any remaining shortfall, the trustee is to pay out the trust
principal. So essentially, the TRU works like a CRUT (without - of course -
the onerous 5, 10, and 50% tests charitable remainder trusts must meet.) We don't
know at this point if the IRS will respect this ordering provision. It should
because it clearly and honestly reflects the payments being made to the current
beneficiary. But if the IRS does not follow this order (as it does in CRTs), a
slight (say 1/4 of a percent) reduction in payout to the current beneficiary may be
necessary. (The ABA's Fiduciary Income Tax Committee has submitted this issue to the IRS
for determination).
• TRUs should typically include the usual discretionary distribution language
( based on ascertainable standards unless the trustee is independent) because, even with
the power and advantages of the TRU, flexibility is appropriate. However, it is not
possible to give the trustee discretion not to pay the distribution amount. The TRU's
strength is predictability. If the beneficiary may not need the income, or the
grantor feels it is important to be able to withhold income from the current beneficiary
for any reason, a fully discretionary trust or the power to adjust may be more
appropriate.
• The TRU is much more than a CRUT - without the charity - or without messy
charitable rules and restrictions. It allows us to think about investing and tax
planning - in a whole new way! For example, there is a very simple but very special
technique which can significantly increase the current beneficiary's after-tax return.
It's called "stock pruning" or "cherry picking" : It's a way to
trim some of the growth while letting the portfolio continue to grow. Stock pruning
is the methodical sale of a small amount of appreciated stock - say 1 to 3% of the
portfolio each year - to help make up the difference between the targeted
distribution pay-out and the actual income produced by trust assets. So if trust
income is insufficient to pay the promised percentage of the last 3 years' average trust
value, the shortfall is made up by selling appreciated stock. But this pruning is
more than a necessity--it's an important tax planning opportunity!
• Although, from a total return investing perspective, "a dollar is a
dollar is a dollar (neither the current nor the remainder beneficiaries care whether it
came from income or principal), from a tax perspective - a dollar of principal will
typically be worth far more than a dollar of income - after taxes.
Picture, for example, a dollar generated from bonds and compare it to a dollar generated
from pruning long term capital gain stocks. The dollar from bonds is subject to
income tax at the highest federal and state rates. The dollar from stocks is
partially capital gain and partially a return of capital. The bottom line is that
pruning results in the current beneficiary netting more - after taxes and much more market
value in the trust for remainder beneficiaries. Using a richer equity mix coupled
with judicious stock pruning, both the current beneficiary and the remaindermen win.
So the trust has both a better overall investment mix and a much better tax result!
Maximizing total return (only possible in a total return type trust) works far
better than the compromise between income and growth that is inevitable with the
traditional trusts we have been drafting!
• Because Code Section1014 eliminates income tax on any appreciation in the
trust, total return investing is useful not only in the family trust where we know we are
going for growth but also in a marital trust where we have traditionally invested heavily
in bonds to try to fill the income needs of the surviving spouse.
• The "ideal payout rate" for a TRU will depend on the client's
goals. The remaindermen will be best served by the lowest pay out to the life
beneficiary. From the current beneficiary's perspective, the payout rate should be
based on the beneficiary's needs and the likely duration of the trust. Our computer
modeling indicates that 3.7 to 4.1 percent work well over long periods in many cases. The
lower the payout rate, the more likely the TRU will accomplish both income and remainder
beneficiary's objectives.
• Why not use a "5or 5" power rather than a TRU? Although it
remains a highly useful tool (even inside a TRU), it remains only a power and has the
following shortcomings: First, it forces the current beneficiary to make a year by year
decision (and the power and extent it is exercised is beyond the control - and to the
detriment of the remainder beneficiary - thus increasing rather than eliminating the
diametric opposition between the two classes of beneficiaries). Second, it is
awkward to administer smoothly since it was not designed to be used continually but was
rather intended to be an escape valve. Third, if the income beneficiary takes too much too
often and at the wrong times (and in addition to trust income), the long term integrity of
the trust is threatened. Fourth, as its name implies, the 5 or 5 provision is a
power and not a system: neither the beneficiaries nor the trustee can plan since there is
no rule governing what will be paid out.
• The most revolutionary and important thing about total return planning and
the TRU is that this new paradigm forces us to ask and with computer modeling - for
the first time we can - realistically analyze: How much income does my client's family
want/and/or need?, and "Are my clients' assets - realistically adequate (after taxes,
trustees' fees, turnover costs, and inflation) - or do they need to be supplemented
(through life insurance or other means) to accomplish my client's objectives?"
Drafting a legally perfect trust is no longer sufficient - if it is not designed to
accomplish this specific client's objectives. Old style trusts are flawed - by
design. They are producing woefully inadequate investment results - despite a
booming stock market. So understanding both what's wrong with current trust design
and how to design them better is essential to every financial services professional.
• Selecting a form of trust should start by considering what trust
design will best match the family's goals. That will often entail mixing or matching of
different types of trusts. We should start thinking about changing our habits of the
past - and use Total Return type trusts (such as the TRU) first, next, the fully
discretionary trust, less often the Indexed pay out trustand lastly choose slowly and
least frequently--the traditional income rule trust! The proper choice will often
not be either the TRU or a discretionary trust but rather a custom designed mix.
Susan Porter of U.S. Trust Company then presented her portion of Special Sesssion I-C:
Susan made four major points:
First, she made it clear she's not in agreement with certain of the provisions in
the recent New York State Legislative Advisory Committee recommendations. These
recommendations would (a) give trustees of existing trusts Section 104 (of the Revised
Principal and Income Act) type power to make adjustments between principal and income and
(b) treat new trusts - by default - as 4% unitrusts unless the governing instrument
proves otherwise. She described the latter proposal as "more of an override than a
default rule" and expressed further concern that the N.Y. proposal, if enacted, would
apply to estates as well as trusts.
Second, she stated that "I have a bias for a fully discretionary trust",
noting that "the trustee can never be criticized unless the trustee was arbitrary or
capricious." She feels strongly that a fully discretionary trust should almost
always be the trust of choice. She cited the strong protection afforded by
discretionary trusts against governmental reach and expressed concern that total return
unitrusts might force more payout of income than income beneficiaries might need.
Third, she expressed the opinion that the D. N. I. of a unitrust doesn't carry out
capital gains.
Fourth, she noted that the N.Y. proposal, defining income as a 4% unitrust interest
(with a three year smoothing rule) presents an issue as to marital deduction
qualification. While she personally felt it qualified, she noted that it is an open issue
which is being discussed with Treasury officials.
She concluded that planners should use unitrusts for predictability and discretionary
trusts for flexibility.
Special Session II-A: Case Study Session on Total Return Trusts, presented by Bob
Wolf and Bruce Guiot:
The general thrust of the session was to display the variety of ways in which Total
Return Trusts can be used to satisfy human needs and investment goals, while helping us
leverage our tax planning and estate planning. By definition, a total return trust is any
trust which allows the trustee to invest for total return and which does not base what the
beneficiary receives on principal and income distinctions. It is not limited to total
return unitrusts, but includes fully discretionary trusts, indexed annuity trusts, total
return unitrusts and other hybrid trusts Real life estate planning situations and
solutions were used to illustrate real families, real problems and real solutions (if a
Total Return Unitrust was used, a TRU Solution).
The initial case study tracked a real trust where the trustee was forced to hold
the principal and pay the income (only) over a 10 year period. Because of a high income
need, the trustee adopted a 50/50 asset allocation, periodically rebalancing to preserve
the allocation and attempt to prop up the income. During the 10 year period chronicled,
while the market value of the trust almost doubled, the actual income distributed to the
beneficiary declined by about 10% because of the decline in interest rates, which
outweighed the dividend increases during the period. The question was then raised: what
should be done with this type of trust which is already irrevocable. The answer was-1)
look for discretionary powers in the instrument; 2) use Section 104 of the UPAIA if it is
in your state, and if not, consider ;3) reforming the trust by court action into a Total
Return Unitrust (TRU).
Key tip in the reformation area is the issue of GST grandfathering. The question is -- if
you reform, might you lose the grandfathered status. The answer is that you may well, but
there is new regulatory guidance under Section 2601 addressing modification. Such
modifications will not cause a loss of grandfathering if:
1. You don't extend the vesting of a GST interest (by another generation, for example).
2. You don't shift a benefit to a lower generation. This raises a strange and intriguing
question, because life estates (such as an income interest) are valued as if the total
return were accounting income and distributed to the beneficiary. That is--the income
beneficiary is assumed to receive a 7.4% rate of return! Hence, even though a conversion
into a 4% unitrust may double the rate the income beneficiary actually
receives, it may be considered to have been decreased-hence violating the rule. ANSWER:
The suggestion on the reformation issue is to make the reformation into an income or 4%
unitrust distribution, whichever is the greater. This should satisfy the rule.
The second case study shows how TRU's can help not only satisfy the NEEDS of the
beneficiary, but also leverage the Applicable Credit Amount. By using a high rate unitrust
(or income, whichever is greater) for the marital and a fully discretionary trust for the
credit shelter trust, the amount remaining after tax for the next generation is increased
over a long period (1960-1998) by 800% over the traditional income model.
One of the most important points of the case studies is that because most
Total Return Trusts try to set a payout, or a payout rate, we are finally dealing with the
one thing the beneficiary REALLY cares about-what they will get. This needs based planning
changes the dynamic of the estate planning process, putting the planner on the same side
with both spouses and children.
The third case study pointed out that in setting a rate for a TRU, the asset
allocation is critical to how much one can pay and still have a rational basis to believe
that the real value of the trust will be preserved. A 2% payout TRU with a 50/50 stock and
bond mix and a 4% payout all equity TRU for most of the period 1960-1998 would have traced
very closely in market value, despite twice as high a payout for the all equity TRU. Moral:
during many periods, you can pay out twice as much with twice the equity exposure (even
though the accounting income may be only half as much)!
The fourth case study involved a second marriage situation where the second spouse
demanded both a high payout from the trust assets and that it be indexed for inflation.
Various approaches were taken to try to match these goals with limited success, at least
when tested by the most difficult investment period for these goals--that is, starting in
1973. Various new types of trust designs were suggested and tested with computer modeling,
including the "no-drop" unitrust, one in which the distribution never goes
down, an indexed payout trust (what was desired, but which didn't work in the worst case
scenarios) and a new hybrid, the TRUCAP Index Trust. This is a trust which is an indexed
payout, but with a cap so that if it pays out more than a set percentage (suggestion was
10%) it becomes a TRU, so as to avoid depletion of the trust, which would occur with a
strict indexed payout. This hybrid seems the best solution for the difficult situation in
which the payout is intended to be indexed for inflation, and illustrates that the process
of inventing new types of trusts to respond to human (not just tax) needs, is just
beginning.
The fifth and last case study illustrated one of the ways in which you can create a
"virtual unitrust" to solve a problem. In that case, the problem was an
overconcentration in international oil stocks with a relatively high yield). By using an
existing withdrawal power, the family could essentially have the benefits of a unitrust,
which could provide the tax advantaged cash-flow (much of the principal pruned to make the
payout is a return of cost basis, and the rest is capital gain), without necessarily
having to simply buy more bonds.
One more point: if people wonder what would be a good second career for trust
lawyers if they really DID repeal the estate tax, how about drafting trusts to meet human
needs!
Several questions were posed as to when the total return unitrust is
contraindicated, and the answers were:
1. Whenever the beneficiary may not need any income from the trust.
2. When the assets of the trust are composed of other than typical financial assets. Real
estate, closely held entities, FLP's, LLC's LLP's etc. don't fit in since illiquid assets,
or ones where you have control over how much the income is, are better held by fully
discretionary trusts.
3. Credit shelter trusts and GST Exempt trusts, to the extent that the grantor doesn't
care about how much is going to be distributed to the beneficiary or to the extent that
the beneficiary doesn't need anything from the trust, work better with fully discretionary
trusts.
As an aside, one of the most useful combinations is to use a TRU in tandem with the
fully discretionary trust. By having a concrete payout from the TRU, often this gives the
beneficiary more confidence to accept a fully discretionary trust for the tax advantaged
trust, because the TRU has satisfied their need for income.
That's it for Report #8.
_________________________________
NOTICE: Although audio tapes of all of the substantive session at the Miami
Institute currently are only made available to Institute registrants for
purchase, the entire proceeding of the Institute are published annually by
Matthew Bender. The text of these proceedings is also available on CD ROM
from Authority by Matthew Bender. For further information, contact your
Matthew Bender sales representative, or call (800) 533-1637, or fax (800)
828-8341, or go to URL <http://www.bender.com/>, or write to Matthew Bender &
Co., Inc., Attn: Fulfillment Dept., 1275 Broadway, Albany, NY 12204.
Back
to Heckerling Table of Contents
|
|
|
|