Some Basic GRAT Concepts
Grantor Retained Annuity Trusts (GRATs), described in Code Sec. 2702(b), are successful in transmitting property out of a property owner’s estate at no estate tax cost and no or minimal gift tax cost to the extent the assets contributed to the trust grow at a rate greater than the so-called Section 7520 rate. The Section 7520 rate is published monthly by the IRS and is used to determine the value of the grantor’s retained annuity in a GRAT and the value of its taxable remainder. Although not certain, it appears that the value of the remainder in a qualifying GRAT can be very small, such as .01 percent of the fair market value of the assets contributed to the trust, if not zero.
Although again not certain, it seems that the annuity payment term of a GRAT may be as short as two years, at which time the annuity payments to the grantor/annuitant cease and anything remaining in the trust is transferred to or for the remainder bene ciaries, such as the grantor’s children, free of estate tax and free of any additional gift tax.
Because it is not certain how small the remainder may be or how short the annuity term may last, WTP has adopted a formula provision created by Diana Zeydel, Esq. of Greenberg Traurig in Miami. Essentially, it provides that the annuity must be at least suf cient to produce the desired remainder sought (such as one percent of the value of the property contributed to the trust) or such larger remainder as is necessary to have a quali ed GRAT, and the annuity term must be the longer of the term desired (such as two years) or such longer term as is necessary to have a quali ed GRAT.
A drafter using WTP can implement the Zeydel Formula on the “Payments During Term” screen in the GRAT dialog, by selecting either the “Fixed Payment Formula” or “Increased Payment Formula” choice under “Select How to Express Annuity,” and then checking the box for “Use the Safe harbor GRAT Formula.”
Short Term Increasing GRATs
Most practitioners agree that “rolling” short term GRATs (such as ones for two years) are preferable to longer ones (such as those for terms longer than two years) for a number of reasons. (“Rolling” means that the annuity payments received by the grantor are contributed to a new short term GRAT.) One of those is that “good” investment performance (e.g., growth in excess of the Section 7520 rate) during the rst two years of a GRAT is not diminished by “poor” investment performance later (e.g., decline during any GRAT period after the rst two years of good performance). And a Monte Carlo simulation that Jonathan and Diana presented with Robert Weiss, a certi ed nancial analyst, at the 2007 Heckerling Institute veri ed that short term rolling GRATs are preferable to longer term ones. As indicated, the study assumed that a “rolling GRAT” approach is used whereby each annuity payment is contributed to a new short term GRAT.
Also, many practitioners have the annuity payments increase from one year to the next. The reason to have the annuity payments increase is that more property stays in the GRAT (than if level annuity payments were made) upon which the growth above the Section 7520 inures to the bene t of the remainder bene ciaries. However, Reg. § 25.2702-3(b) provides that annual increases in annuity payments above 20% are disregarded for purposes of valuing the annuity interest retained by the grantor and, therefore, the value of the gift of the taxable remainder. Accordingly, the annuity payments in a GRAT should not increase by more than 20% a year.
New “Science”: Short Term Declining Rolling GRATs
Although it may seem counterintuitive, from both a theoretical perspective and based upon indications Monte Carlo simulations, it seems that a steeply declining short term rolling GRAT is better than a level payment or increasing one. The reason is that, if the performance in the rst year is good (i.e., growth in excess of the Section 7520 rate), the extra growth more likely can be “locked” in for the remainder bene ciaries by paying the grantor off almost in full at the end of the rst year (even if investment performance is poor in the second year). And, if the performance in the rst year is poor (e.g., a steep decline in the value of the assets contributed to the trust), all (or nearly all) of what is in the GRAT is returned to the grantor who can “re-GRAT” the assets returned (that is, create a new GRAT with the assets that have been distributed as the rst annuity payment). In fact, even if the performance in the rst year is excellent, assets returned to the grantor as the rst year annuity payment should be “re-GRAT-ed” for the remainder bene ciaries. And although, as stated, the regulations effectively limit how much an annuity can increase from one year to the next, they do not limit how much it can decline.
Using the basic Zeydel Formula as a model, we have added another choice under “Select How to Express Annuity”, which new choice is “Declining annuity payments.” If you make that selection, you will be asked to specify the percentage the rst year annuity will be of the initial fair market value of the annuity contributed to the trust. For example, if it is a two year GRAT, you might provide that the rst year annuity payment will be 90% of the initial fair market value of the assets contributed. Under the new ”Zeydel” formula, the annuity payment for the rst year will be the lesser of 90% or the largest percentage the annuity can be in the rst year and be a quali ed GRAT. The annuity payment for the next and nal year of the GRAT will be determined by a word formula so the remainder will be the greater of (1) the size desired (e.g., .01 percent of the initial fair market value of the assets contributed to the trust) or (2) the minimum size a remainder may be in a quali ed GRAT.
Summary and Conclusion
Both theory and Monte Carlo simulations suggest that a steeply declining GRAT is more likely to be successful in transmitting wealth than a level payment or increasing annuity payment GRAT. WTP now offers a steeply declining GRAT.
InterActive Legal, LLC 6767 N Wickham Rd Ste 305 Melbourne, FL 32940 321-252-0100 firstname.lastname@example.org
– © 2008. Jonathan G. Blattmachr, Michael L. Graham & Diana S. C. Zeydel All Rights Reserved.