Current legislation passed by the House and being considered by the Senate Finance Committee would restrict the use of grantor retained annuity trusts (GRATs) by requiring that they have at least a 10-year term and that the amount of annuity payments provided for not decline from year to year. These requirements would eliminate “rolling” two-year GRAT planning funded with marketable stock, which has grown to become an increasingly popular, as well as effective, wealth transfer strategy. The proposed legislation would also increase the likelihood that GRAT assets could be included in a grantor’s estate given the higher probability of the grantor’s death during the longer minimum GRAT term.

The prospect of the legislation’s final passage raises a number of important future planning issues for donors. But are there GRAT (and other) planning opportunities now, in advance of the possible legislative change, that donors might consider pursuing? While there is no way to know for sure, all indications are that the proposed legislation will be effective only upon enactment. Thus, until a bill passes the Senate and is signed by the President it should be possible to establish GRATs with terms shorter than 10 years.¹

Using our Wealth Forecasting System, we looked at different scenarios to determine the best advice regarding GRAT planning as well as alternative wealth transfer strategies, including a loan or installment sale to an intentionally defective grantor trust (IDGT). Our research indicates that there are several key factors to consider in the decision, including the donor’s time horizon or expected mortality, the level of interest rates when the strategy is established, the timing of payments to both grantor and beneficiaries and the ability to impact the amount of wealth transferred by scaling up the amount of capital committed to the strategy.

Mortality Risk

First of all, a longer GRAT term brings with it higher mortality risk—risk that the grantor may die during the term, and the resultant exposure of GRAT assets to taxation in the grantor’s estate. This is arguably the most consequential factor in the GRAT decision. A donor age 65 has the prospect of a “long planning horizon.” His mortality risk over the next decade is relatively low, with an almost 85 percent chance of surviving a GRAT of 10 years in duration (See Display 1). But the likelihood that a male age 75 would survive beyond the 10-year term is just 63 percent, and the probability of an 85-year-old outliving a GRAT established post-legislation is less than one in three.  

Display 1:
*Based on RP-2000 mortality tables, single, male.
Source: AllianceBernstein

 

So how might donors with different time horizons think about their transfer options, given what may be a fleeting pre-legislation planning opportunity? Someone with a reasonably long time horizon (that is, one of at least 12 years in length) might consider establishing a pre-legislation GRAT with a two-year term funded with marketable stock to capture the benefit from a one time “roll” of the GRAT annuity payments to two subsequently established 10-year term GRATs. Alternatively, a donor with a more limited time horizon might consider establishing a GRAT with the longest term consistent with his mortality risk—of less than 10 but more than two years. To remove the question of mortality risk altogether, one might consider alternative strategies, such as a loan or installment sale. The key advantage of those strategies is the absence of mortality risk, where only unpaid principal and interest is subject to estate tax at the grantor’s death. But there are other factors to consider besides mortality.

Interest Rates

Today’s low interest rate environment is another key planning factor. Our prior research showed that rolling short-term GRATs are a superior strategy to longer term GRATs (and most other alternative strategies, including installment sales) in all interest rate environments (See David L. Weinreb and Gregory D. Singer, “Rolling Short Term GRATs are (Almost) Always Best,” Trusts & Estates, August 2008). However, longer term GRATs (as well as loans and installment sales) are more likely to succeed—and are more attractive to planners—when transfer hurdle rates are lower, as they can in effect “lock in” the lower rate.² Although the Internal Revenue Code Section 7520 rate and applicable federal rates (AFRs) have risen from their historic lows in February 2009, they remain extremely attractive. That gives longer term GRATs an advantage today, as shorter term GRATs have higher interest rate risk—greater exposure to higher IRC Section 7520 hurdle rates when future GRATs are to be established. Using our Wealth Forecasting System, we project rising rates going forward, with only 16 percent of our 10,000 trials resulting in an IRC Section 7520 rate less than (today’s low) 3.4 percent 10 years from now.³

To give a sense of the impact of the rate environment on the success of longer term GRATs, we considered how 10-year GRATs would have performed under different historical conditions. The results are significant (See Display 2): The inflation-adjusted median remainder to beneficiaries from 10-year GRATs established with $1 million in a globally diversified equity portfolio in low IRC Section 7520 rate years (the lowest quartile, 1.2 percent to 3 percent) is $1.9 million, more than three times that of such GRATs established in high IRC Section 7520 rate years (the highest quartile, 9.2 percent to 19.4 percent).⁴

Display 2:
*Over 684 10-year periods, beginning monthly from 1941-1998; using a proxy for the IRC Section 7520 rate before 1989 based on IRS methodology.
**"Low rates" refers to the lowest quartile of initial 7520 rates, ranging from 1.2 percent to 3.0 percent. "High rates" refers to the highest quartile of initial IRC Section 7520 rates, ranging from 9.2 percent to 19.4 percent. All strategies funded with $1 million. All assets are invested in a globally diversified portfolio composed of 35 percent U.S. value stocks, 35 percent U.S. growth stocks, 25 percent developed country international stocks and 5 percent emerging market stocks. Wealth to beneficiaries is reinvested and adjusted for inflation.
Source: AllianceBernstein

 

A loan or installment sale to an IDGT can also benefit by locking in today’s low rates (the relevant AFR), which is lower than the IRC Section 7520 rate in the case of a loan or purchase note term of nine years or less. But unlike a GRAT, a loan or installment sale is subject to market risk—risk that assets purchased by the IDGT may underperform the AFR, eroding the value of the IDGT seed gift or prior funding.

The Long and Short of the Pre-Legislation Transfer Decision

We analyzed the possibility of capitalizing on the remaining time before any legislative action by funding a two-year GRAT with marketable stock now. According to our forecasts, establishing such a GRAT will offer only a modest benefit from the ability to “roll” two annuity payments to 10-year post-legislation GRATs,⁵ unless the property contributed to the GRAT is so significant in value that even a relatively small outperformance might have meaningful impact. However, committing some funding to a two-year GRAT prior to a potential legislative change would provide a hedge against the possibility that the legislation in fact might not be enacted, leaving rolling two-year GRATs the superior planning alternative. Short-term GRATs established prior to the enactment of legislation also remain appealing in “opportunistic” cases: for example, if funded with pre-IPO stock or stock in a private company valued at a discount, both of which would benefit from the increase in value related to the marketability of the asset upon a completed transaction.

What about pre-legislation intermediate-term GRATs, or even long-term GRATs, pre- or post-legislation? Let’s assume that a donor is considering the alternatives of committing $1 million to a four-, seven-, 10-, or 15-year GRAT invested 100 percent in global stocks (See Display 3). The probability of a remainder greater than $0 ranges from 70 percent for the four-year GRAT to 90 percent for the 15-year GRAT. If each remainder is held through year 15 in an IDGT that is also invested 100 percent in global stocks, the inflation-adjusted median wealth transferred projects to range from $247,000 for the four-year GRAT to $881,000 for the 15-year GRAT. ⁶

Display 3:
*Each GRAT established when IRC Section 7520 rate was 3.4 percent and and assumed to have 20 percent increasing term annuity payments. Each GRAT established with an initial commitment of $1 million. The GRAT remainders are invested in an IDGT through year 15 of the analysis. Global stocks are 35 percent U.S. value/35 percent U.S. growth/25 percent developed international/5 percent emerging markets. Based on Bernstein's estimates of the range of returns for the applicable capital markets over the next 15 years. Data don't represent any past performance and aren't a promise of actual future results. 

 
 

Other Alternatives: A Loan or Installment Sale to an IDGT

A donor might also consider a loan or an installment sale to an IDGT as an alternative strategy separate from or in combination with GRATs. As we’ve mentioned, this strategy has advantages (including reduction of mortality risk and the ability to lock in current low AFRs) as well as disadvantages (the IDGT borrower, or the installment note purchaser, bears market risk that its investments might fail to outperform the AFR). To see how they compare, we looked at the range of wealth transferred by three strategies, each funded with $1 million, plus $100,000 that is contributed to an IDGT in each case so as to ensure their comparability, since the loan or installment sale to an IDGT is assumed to require seed funding (See Display 4).⁷  In the case of the “staggered vintage GRATs” we assume that 25 percent of the $1 million is allocated to each of a four-, seven-, 10-, and 15-year GRAT. The proceeds of each strategy are held in the IDGT, invested 100 percent in global stocks through the end of the 15th year following the implementation of each strategy.

The loan/installment sale is the best performer in typical markets, but there’s approximately a 10 percent chance that it will result in no wealth transferred at all. That means a loss of the entire $100,000 seed gift.⁸ Also important, note how closely the staggered term GRATs compare with the 10-year term GRAT, and yet mortality risk is reduced by the staggered terms.

Display 4:
*A loan in the amount of $1 million is made for a nine-year term, annual interest only and a balloon principal payment. IDGT to which loan made is funded with a seed gift of $100,000. A mid-term AFR of 2.87 percent applies to the loan. Any value remaining after full loan payment is held in the IDGT for the remaining six years of the analysis. The staggered vintage term GRATs are comprised of a four-year, seven-year, 10-year and 15-year GRAT, each with an initial value of $250,000. The applicable IRC Section 7520 rate is 3.4 percent and each GRAT  has 20 percent increasing annuity payments. The GRAT remainders are invested in an IDGT through year 15 of the analysis. One hundred percent global stocks held in all accounts. Based on Bernstein's estimates of the range of returns for the applicable capital markets over the next 15 years. Data don't represent any past performance and aren't a promise of actual future results.     

 


Also consider the timing of beneficiary access to the funds, as well as the pace of annuity payments to the grantor. Both are later in the case of longer term GRATs. Staggered term GRATs compare well with a 10-year term GRAT in terms of total wealth transferred, but funds above the initial $100,000 gift to the IDGT can be made available to beneficiaries from the staggered GRATs beginning after year 4 (assuming the four-year GRAT succeeds in transferring some amount to the remainder beneficiary), with rising amounts available as other successful intermediate GRAT terms expire. And the grantor also accesses annuity payments faster in earlier years from the staggered term GRATs. With alternative strategies, like an installment sale, beneficiary access to trust assets and lender payment is also more flexible than 10-year term GRATs, but the amounts are more dependent on investment performance.

Scalability

Finally, it’s important to note that a distinguishing characteristic of a “near zeroed-out” GRAT, whatever its term, is that it can be “scaled” without gift tax consequences if additional assets are available. If a donor would like to improve the probability of meeting his target, he could commit more assets to the strategy. On the other hand, a loan or installment sale to an IDGT must be limited in size to a multiple of the seed gift made to a newly established IDGT, or the prior net worth of a “seasoned” IDGT, for the transaction to avoid challenge. Such strategies are limited in their ability to “scale” the commitment of capital, unless the donor is willing to incur gift tax to make a larger seed gift to the IDGT to support a larger loan or installment sale.⁹ Because “near-zeroed-out” GRAT planning can be undertaken with only insignificant gift tax exposure, even if no remaining lifetime gift exclusion is available, it can be used in conjunction with other planning that uses lifetime gift exclusion to improve the probability of achieving wealth transfer goals.

Provisional Conclusion for Pre-Legislation Planning

In sum, grantor mortality risk is arguably the primary and therefore governing risk in this decision, so we can organize our advice around that factor. In the case of a grantor with high mortality risk, consider establishing a pre-legislation GRAT with rising annuity payments and the longest term consistent with mortality risk management. This way the donor can lock in a low hurdle rate, reduce market risk and can scale up the amount contributed to the GRAT. Donors might also consider alternative strategies such as a loan or installment sale to an IDGT so as to lock in a low AFR and reduce mortality risk, though they will still be exposed to market risk.

Grantors with low mortality risk (say 55 years of age or less) should consider longer term or staggered vintage GRAT(s) to lock in the current low hurdle rate and avoid interest rate risk; they should establish them pre-legislation if less than a 10-year term is preferred for earlier beneficiary access or otherwise. Alternatively, they could consider establishing a pre-legislation GRAT with a two-year term to capture the benefit from the onetime “roll” of the GRAT annuity payments to two subsequently established 10-year term GRATs. In both cases, these donors could scale the GRAT to increase the magnitude of their wealth transfer. They could also consider combining such an approach with alternative planning strategies (such as a loan or installment sale to an IDGT) to lock in today’s low AFRs and reduce mortality risk, recognizing that market risk is a factor in such planning that can be avoided by using GRATs.

Endnotes

 
1.    “Near zeroed-out” grantor retained annuity trusts (GRATs) appear likely to remain available. GRATs are “zeroed out” by setting annuity payments at levels such that their discounted present value, applying the Internal Revenue Code Section 7520 rate, is approximately equal to the value of the property committed to the GRAT. As a result, GRATs of any size may continue to be established without making gifts of any significance.
2.    Furthermore, with rising, or so-called back-loaded annuity payments, more assets are kept working in the strategy longer.
3.    Our Wealth Forecasting System projects 10,000 paths of returns for a wide range of asset classes. Our forecasts (1) are based on the building blocks of asset returns, such as inflation, yields, yield spreads, stock earnings and price multiples; (2) incorporate the linkages that exist between various asset classes; (3) take into account current market conditions; and (4) factor in a reasonable degree of randomness and unpredictability.
4.    Based on 684 10-year periods, commencing monthly, from 1941–1998.
5.    We compared the following strategies and found a small advantage to establishing a single short-term GRAT funded with marketable stock in advance of the pending legislation. Scenario 1: $1 million in global stocks contributed to a two-year near zeroed-out GRAT. On the first anniversary, roll the two-year GRAT’s first annuity payment into a new 10-year zeroed-out GRAT. On the second anniversary, roll the two-year GRAT’s second annuity payment and the 10-year GRAT’s first annuity into a second new 10-year zeroed-out GRAT. At the end of years 2, 11, and 12, GRAT remainders are contributed to a grantor trust, also invested 100 percent in global stocks. Scenario 2: $1 million in global stocks contributed to a 10-year zeroed-out GRAT. On the first anniversary, roll the 10-year GRAT’s first annuity payment into a second new 10-year zeroed-out GRAT. On the second anniversary, roll the annuity payments from each 10-year GRAT into a third new 10-year zeroed-out GRAT. At the end of years 10, 11, and 12, GRAT remainders are contributed to a grantor trust, also invested 100 percent in global stocks. Initial IRC Section 7520 rate is 3.4 percent. Each GRAT is assumed to have 20 percent increasing term annuity payments. Global stocks are 35 percent value/35 percent U.S. growth/25 percent developed international / 5 percent emerging markets. There was a 62 percent probability of an equal or better wealth transfer outcome in Case 1 versus Case 2, with a median net increase in wealth transferred in Case 1 of approximately $30,000 (3 percent of the amount committed to each strategy). This analysis is based on Bernstein’s estimates of the range of returns for the applicable capital markets over the next 12 years.
6.    The inflation-adjusted median remainder ranges from $151,000 for the four-year GRAT to $881,000 for the 15-year GRAT.
7.    A seed gift of 10 percent of the amount of the loan or the value of the assets to be purchased appears to be the general rule under prevailing practice. However, no tax or legal authority expressly sanctions 10 percent as a necessary or sufficient amount.
8.    There is an 87 percent likelihood that an inflation-adjusted $100,000 will be transferred at the end of 15 years.
9.    It’s worth noting that, given the reduced gift tax rate for 2010 of 35 percent, the appeal for some families of making taxable gifts is dramatically heightened, particularly considering the 55 percent gift tax rate that would apply in 2011 and afterward, pending any estate tax legislation. For more on this opportunity, see our recent white paper, Investment Opportunity amid Tax Uncertainty.