Letting GRATs Fail for Success

If a 2-year grantor-retained annuity trust (GRAT) is to be successful at transferring wealth to its remainder beneficiaries, it must achieve significant positive returns in its first year; otherwise, its chances of success are staggeringly improbable. Indeed, in the last decade, most GRATs likely transferred significant amounts of wealth, considering that the U.S. stock market has seen positive returns in nine out of the last 10 years, averaging an annualized 10.7 percent return over that period. That steep and steady growth in U.S. equities, coupled with a period of ultra-low interest rates, has been a boon for clients with GRATs. However, perfect GRAT conditions can’t continue indefinitely, and it stands to reason that they’ll be more challenged moving forward. Threatening conditions can take the form of rising interest rates, as we’re beginning to see now, or increasing downside volatility, which is significant for GRATs, where success and failure is quite often a byproduct of timing. While we don’t pretend to know how to time the equity markets, we do know when you should pull the plug on your GRAT.

Two Choices

The grantor of an underwater GRAT (that is, a GRAT that’s flat or negative compared to its funding value) has two choices. She can either let it ride in the hopes that its underlying assets appreciate rapidly enough in the time remaining to make it successful, or she can cut her losses before the first annuity date and attempt to capture any subsequent appreciation on those assets in a new GRAT (by exercising her swap power or using a note). Which choice a grantor makes will depend on her assumptions about the underlying asset’s appreciation potential relative to the magnitude of the decline of the GRAT’s underlying assets. If a grantor knew with any certainty that her GRAT would fail, she could take steps to accelerate and minimize that failure before the first annuity payment by creating a new GRAT with the old GRAT’s assets (after a swap or sale). But, when does certainty arise? If the value of the GRAT has declined by 50 percent from the time it was funded, any practitioner might encourage the grantor to cut her losses and institute a new GRAT, especially if the grantor believed in the assets’ appreciation potential going forward. But, what if the GRAT’s value has declined by 10 percent from its initial funding value? What about 5 percent? 2 percent? What if it appreciated, but only by 1 percent? 

Using quantitative analysis, we’ll show that the grantor’s choice is almost always clear: If her 2-year GRAT¹ is funded with marketable securities and is underwater at all prior to the first annuity date, she should cut her losses and start a new GRAT. That is, no matter the Internal Revenue Code Section 7520 rate, even if her GRAT is underwater by just 2 percent prior to the first annuity date, her GRAT will almost certainly fail, and, even in the unlikely event that it doesn’t fail, re-GRATing yields superior results in nearly all instances.

The Mathematics of Failure

If a GRAT closes out its first year underwater, the chances of the GRAT successfully transferring any wealth to the remainder beneficiaries become statistically improbable. To understand why a GRAT, once underwater, has a difficult time achieving success, consider first some basic mathematics of percentages. Suppose a stock worth $100 per share declines 20 percent to $80 per share. To make it back to $100, the value of the stock must appreciate not by the 20 percent it lost (that would leave just $96) but by 25 percent; if the price dropped from $100 to $75, a 25 percent decline, the stock must appreciate by 33 percent; a decline of 50 percent requires 100 percent appreciation to be made whole, and so on. In short, more substantial returns are required to recoup losses. A GRAT exacerbates this principle, because approximately half of the GRAT’s assets are distributed at the 1-year mark, leaving a much smaller pool of assets to recoup any losses. For example, suppose a $100 stock drops 20 percent to $80, and then half of the position is sold. A 50 percent return (not 25 percent) is required on your client’s remaining $40 to recoup her initial investment. The same math applies to GRATs.

Example: 2-Year GRAT Established

To illustrate how this works, suppose that Grayson establishes a 2-year GRAT when the IRC Section 7520 rate is 3.4 percent and funds it with an S&P 500 exchange-traded fund, valued at $1 million. The annuity payments are varied, such that the second annuity is 20 percent greater than the first, resulting in two required annuity payments of $478,583 and $574,299. Just prior to the first annuity being paid, Grayson learns that the GRAT’s investments have declined by merely 2.5 percent to $975,000. At this point, the GRAT pays Grayson his first annuity of $478,583, leaving just $496,417 in the GRAT. For Grayson’s GRAT to transfer any wealth to the remainder beneficiaries, the Year 2 returns must exceed 15.7 percent, no small feat for a market that has a historical average annual rate of return of just under 10 percent. Indeed, to leave just $50,000—5 percent of the initial funding value of Grayson’s GRAT—to the remainder beneficiaries, a 25.7 percent return in Year 2 is required. The historical probability of such a return in any given year is just 10.75 percent. That is, while Grayson’s GRAT was down a mere 2.5 percent in Year 1, it has just about a one in 10 chance—based on historical returns—of transferring a small amount of wealth to his remainder beneficiaries. 

Breaking Even

To showcase just how difficult it is to achieve a successful GRAT following negative Year 1 returns, we’ve plotted the relationship, represented by the black line on “Required Year 2 Returns Given Year 1 Performance,” p. 25, between assumed first year returns and second year returns needed after the first annuity payment is made for Grayson’s GRAT to break even; that is, for the trust to have enough assets to pay the final annuity payment without any left to transfer to the remainder beneficiaries. Any amount above this black line represents a successful transfer of wealth. For reference, the red line illustrates the returns necessary to achieve a remainder of just 5 percent of the GRAT’s initial funding value. We’ve assumed throughout this article that each GRAT is zeroed-out and that the second annuity payment is equal to 120 percent of the first.

Adding the S&P 500 historical average, represented by the red dotted line, highlights how any negative return requires an above-average rate of return merely to break even, let alone achieve a successful wealth transfer. In fact, even smaller positive returns in Year 1 can require above-market average returns in Year 2 to successfully transfer any wealth. In addition, we’ve translated the returns necessary to break even and to produce a 5 percent successful wealth transfer into probabilities based on historical S&P 500 returns in “Likelihood of Grantor-Retained Annuity Trust Success Given Year 1 Returns,” p. 26.² For example, if Grayson’s GRAT declined in value by 10 percent in Year 1, the Year 2return needed just to break even is a staggering 36.3 percent, the probability of such an event being just 2.09 percent. Stunningly, the probability of successfully transferring 5 percent of the GRAT’s initial funding value to the remainder beneficiaries in this scenario is virtually zero.

Success for Market Sectors

Because it’s atypical for a GRAT to be invested in the very diversified S&P 500, we also ran the same historical analysis for Grayson’s GRAT assuming it had instead been invested in individual industry sectors, such as consumer staples, financials, healthcare and information technology stocks within the S&P 500. We’ve plotted the results in “Probability of Success for Various Market Sectors Given Year 1 Returns,” p. 26, which shows the relationship, specific to those industry sectors, between the Year 1 returns and the probability of the GRAT transferring at least 5 percent of its initial funding value to the remainder beneficiaries. The results paint a similarly bleak picture for a GRAT’s chance of success. For example, even when the GRAT has increased in value by 1 percent on its first anniversary, there’s no industry where that results in even a 50-50 chance of a successful outcome. 

We believe that these numbers showcase the futility in continuing a 2-year GRAT when there are losses just prior to the first annuity payment. In fact, even when the Year 1 returns are flat or positive (but below the Section 7520 rate), the necessary Year 2 returns prove to be challenging. To be sure, we took the same historical data and ran it through a financial forecasting model that simulated 15,000 market scenarios for each investment given our Year 1 return/loss assumptions. The results were consistently grim. When Grayson’s GRAT is down 2.5 percent in Year 1, our model produced a break-even result in less than 32 percent of outcomes and a successful result (measured again by 5 percent yield to remainder beneficiaries) in only 11 percent of outcomes. Of the 32 percent of outcomes in which the GRAT managed to break even, the average remainder was $42,000. Even if the GRAT was invested in information technology sector stocks within the S&P 500, the best performing individual asset class in our forecasting models, the GRAT broke even in less than 42 percent of outcomes, and success occurred in only 23 percent.  

Effect of Rising Interest Rates 

Rising interest rates add weight to the sinking ship. Because a higher Section 7520 rate means larger annuity payouts, the principal remaining to generate a larger return in Year 2 is smaller, meaning the returns in Year 2 must be even greater to compensate for poor performance in Year 1. That is, as interest rates rise, the likelihood of a successful GRAT after Year 1 losses becomes increasingly and vanishingly small. To show the effect, consider again Grayson’s GRAT that declined in value by 2.5 percent (or $25,000) in Year 1. Recall that to break even in Year 2, the investments would need to appreciate at a rate of 15.7 percent, and to transfer just 5 percent of the initial funding value to the remainder beneficiaries, the assets would need to appreciate at a rate of 25.8 percent. Now suppose that Grayson funded his GRAT when the Section 7520 rate was 5.8 percent (the historical average excluding the most recent nine years during which interest rates have been artificially low). After the initial 2.5 percent decline in Year 1, the Year 2 return required to break even increases to a difficult-to-achieve 24.1 percent, and the Year 2 return required to transfer just $50,000 to his remainder beneficiaries climbs to an improbable 34.6 percent. Our forecast models yielded similarly dire outcomes at this rate, breaking even in less than 15 percent of outcomes. That is, when Grayson’s GRAT declined in value by just 2.5 percent in Year 1, the GRAT failed 85 percent of the time. “Effects of Rising Rates on Welfare of Grantor-Retained Annuity Trust,” this page, compares the break-even outcomes at various Year 1 returns for a Section 7520 rate of 3.4 percent, 4.2 percent and 5.8 percent. 

Notably, as interest rates rise, even small but positive Year 1 returns that are less than the Section 7520 rate require increasingly substantial Year 2 returns. For example, when the Section 7520 rate is 5.8 percent, a positive return of 2 percent in Year 1 requires above-average market returns in Year 2 just to break even. The return necessary to create a successful wealth transfer of just 5 percent is a staggering 23 percent. Historically, the likelihood of that return is about 15 percent, a surprisingly low probability given the positive return in Year 1 in this instance. 

The Effect of Re-GRATing

When presented with a decline in value of his GRAT’s investments in Year 1, Grayson believed he had two choices: (1) let it ride and hope for a successful outcome, or (2) freeze the GRAT and use the assets to capture any upside potential in a new GRAT. The foregoing shows that Grayson in fact has only one rational choice, because his hope for a successful outcome is sobered by the significantly low probabilities of those outcomes. Indeed, embedded in the hope for a successful outcome in the former option is a belief that the assets will perform significantly better in their final year. Thus, if Grayson believes the assets have appreciation potential, and if he’s retained a swap power, he can reacquire the assets by substituting cash or other assets of equivalent value and transferring the assets to a new GRAT to capture any potential appreciation. Alternatively, the trustee could sell the assets at their fair market value for cash or a promissory note, and the grantor could use those assets to fund a new GRAT.³ 

For Grayson’s GRAT, we highlighted the mathematics of failure after a 2.5 percent loss in the first year. When we inputted the assumptions for Grayson’s GRAT into our forecasting models, breaking even occurred in only 31.4 percent of our 15,000 simulated outcomes. The Year 2 rate of return, on average, in those break-even outcomes was 24.2 percent, and the remainder averaged $42,262, less than 5 percent of the funding amount. If Grayson had in fact decided that the math was against him and froze the GRAT and used the assets to fund a new one (assuming the same Section 7520 rate), a 24.2 percent return in Year 1 of the new GRAT now creates a successful outcome in 99.5 percent of simulated outcomes, averaging a remainder of over $260,000! In short, the outcome from re-GRATing is far more favorable than any possible outcome from letting it run its course and hoping for a successful wealth transfer. If a large return is required to dig the GRAT out of its hole, the return should be earned inside a new GRAT, plain and simple. 

Turn Failure Into Success

Practitioners have an opportunity here. Rather than shrugging off small declines in value of their clients’ GRATs and counseling them to let the GRATs ride and hope for a large enough upswing in the market to produce a successful wealth transfer, practitioners can help their clients turn inevitable failure into great success. As we’ve shown, even very small downturns (and even small positive returns!) generally require massive, outsized upswings just to break even, let alone produce a successful wealth transfer. Once the first annuity payment is made after a very small decline, historically speaking, a statistical anomaly is required to produce a successful outcome. And, even if that successful outcome is achieved, it will always make sense to achieve that outcome in a new GRAT. Practitioners should routinely monitor investment performance of GRATs just prior to the first annuity date, and if the performance is down at all, or even flat, the assets (especially if they are publicly-traded equities) should be frozen or sold and re-GRATed. 

Endnotes

1. The content of this article exclusively focuses on rolling 2-year grantor-retained annuity trust (GRAT) strategies. Much has been previously written on the benefits of the rolling 2-year strategy relative to the longer-term strategies regardless of the interest rate environment. See, for example, David L. Weinreb and Gregory D. Singer, “Rolling Short-Term GRATs Are (Almost) Always Best,” Trusts & Estates (August 2008). In addition, we believe that predictions for short-term asset price performance and the longer term trajectory of monetary policy shouldn’t be used to influence the duration of a GRAT, as this may not lead to favorable or repeatable outcomes. Separate analyses should be done when considering whether a GRAT with a longer term than two years might fail by a particular annuity date.

2. To accomplish this, we took the returns and annualized volatility of the S&P 500 dating back to 1929 to create a normally distributed bell curve, which then allowed us to calculate the probability for a given 1-year return.

3. In addition, if the GRAT has truly vested beneficiaries, those beneficiaries could potentially disclaim their interests prior to the 9-month anniversary. While our numbers focus on the 1-year anniversary, since the 1-year mark is when the first annuity locks in a loss, if the GRAT’s assets have sufficiently underperformed, it may make sense at the 1-month anniversary, as well. One should be certain, however, that there are no estate inclusion issues if the grantor holds a reversionary interest in the GRAT’s assets.