National Association of Estate Planners and Councils

December, 2007 Newsletter
Provided by Leimberg Information Services

See other issues.

Jelke Revisited


We received a number of comments on the Jelke case appeal we reported as Estate Planning Newsletter # 1205 (Nov 19, 2007).  You'll find varying views and fascinating viewpoints from Jeff Pennell and Technical Editor Steve Gorin as well as LISI Commentator Team members Paul Hood and Mike Jones.


Advisors have argued successfully for valuation discounts for the income tax liability that flows with certain assets (the exception being the income tax attributable to a right to receive income in respect of a decedent, such as an inherited IRA or qualified plan, for which the §691(c) deduction instead is available). Thus, for example, taxpayers have succeeded in generating a valuation discount for built-in capital gain, as allowed in cases such as Eisenberg, Davis, and  Jameson.

These decisions ultimately concluded (and the government ultimately conceded) that

[t]he value of the stock of a closely held investment . . . holding company . . . is closely related to the value of the assets underlying the stock . . . and the cost of liquidating it.

They resulted in valuation discounts to reflect unrealized capital gain in corporate assets, even if no liquidation or sale of assets was contemplated. It is well agreed now that a willing seller and a willing buyer would trade at a lower price than if there was no built-in capital gain liability. According to Davis, there is a less ready market for assets with tax liabilities, which Davis considered in establishing the appropriate lack of marketability discount.

The tax liability in Davis was just one of several factors that informed the discount: not the full amount of that tax but a discount that reflects the likelihood of a willing buyer incurring the tax, and when. The Davis discount was 15%, attributable to an unrealized capital gains tax liability that the court indicated would exceed 37% for state and federal purposes combined. That discount was roughly one-third of the aggregate marketability discount allowed overall.

According to Eisenberg, a hypothetical willing buyer definitely would consider the liability as one of many factors in making a valuation of the property, and likely would pay less because of the contingent tax liability. Explicit was the notion that a dollar-for-dollar reduction in value was not proper, the Eisenberg court pointing out that, among a universe of potential willing buyers, some might conclude that the tax liability was a lesser concern than others, because of differences in their plans for the stock or the corporation and its underlying assets, along with the tax attributes of those various buyers.

Estate of Dunn involved essentially the same issue as in Eisenberg and Davis — only better (in terms of the tax liability and therefore the discount) — and cast doubt on the best method for handling these types of cases. Dunn involved built-in capital gain tax liability that was attributable to accelerated depreciation and therefore would be recaptured at ordinary income tax rates, so the liability was not a capital gains tax but an ordinary corporate income tax. The taxpayer valued the business itself in two ways — using a net asset approach in which this built-in tax liability was relevant, and a cash flow approach in which it was not — and the court wrestled with the question of how to reflect those two values in determining the value of the business entity itself.

One issue the court on appeal had in reversing the Tax Court was the relative weight it should give to the two valuation approaches, the court ultimately deciding that the cash flow approach was the more probative because the court expected that a buyer would purchase the business for its going concern value.

That weighting issue was significant because the court on appeal found that the proper approach to considering the built-in income tax liability was at the level of valuing the entity, not at the Eisenberg/Davis level of taking the value of the entity (determined without this factor) and then reflecting the built-in tax liability when determining the size of the discount for marketability that a willing buyer would consider in deciding how much to pay for the taxpayer's interest in the entity.

That distinction is important because, when comparing various valuation approaches, it is entirely possible that reflecting the tax liability in the net asset analysis would yield a larger ultimate tax saving than reflecting it as one of several factors that increase the percentage marketability discount applied after the entity value is established.

The important lesson of Dunn, then, is that taxpayers may have options in choosing how or when to claim built-in tax liabilities in the valuation process.

Into this valuation environment came Jelke, in which the government argued (successfully at the Tax Court) that any discount from net asset value for built-in capital gain should be time-value-adjusted to reflect the expected delay in liquidation of appreciated assets and therefore the delay in imposition of the built-in tax liability.

On appeal that time-adjusted approach was totally rejected, the court instead presuming liquidation and immediate payment of all the tax on appreciation as of the date of death. Involving a C Corporation that had sold off its operating assets long before the decedent died and was simply a holding company for marketable portfolio investment assets (with stipulated values), the only issues were proper tax affecting for the built-in tax liability and the proper lack of control and lack of marketability discounts for the decedent's slightly greater than 6% interest in the entity. On the latter two matters the discounts were quite small — 10% and 15% respectively (with nary a mention of whether either discount was appropriate if the court was assuming total liquidation of the entity and immediate distribution of the proceeds from selling all the underlying assets).

Regarding the built-in gain discount, the Tax Court concluded that over $51 million of gain tax should produce a reduction in value of only $21 million. On appeal the full $51 million tax liability was allowed as a dollar-for-dollar valuation reduction.

Interesting about Jelke are a number of issues that arise in the determination of the proper calculation of the effect of the built-in income tax liability and the time-value issue itself. A financial or investment analyst might properly explain that gain and tax issues are reflected in the fair market value of each asset for purposes of a willing-buyer, willing-seller assessment. Indeed, ignoring for a moment any past gain (which is unavoidable in this situation), all future gain is an implicit factor in valuing any asset, in the sense that investors always consider the income and appreciation potential, and the tax character of both, when determining whether to buy an asset for future investment performance.

The time when any income, gain, or tax will be generated also will be reflected automatically in the market value of a publicly traded asset like the underlying investments in the Jelke C Corporation. So, perhaps looking at gain that comes built in ought to be considered in the same vein.

That was not the court's approach in Jelke, however, probably because it was not the parties' approach. Indeed, if there are three steps in the valuation process with this holding company, it is at the first level that the tax affect might best be reflected but it has been at the other two steps that the case law to date has wrestled with it.

That is, the court in Jelke did not (nor did other courts before it)

  1. determine the value of each asset, with its built-in tax liability as one factor, then

  2. etermine the value of the C Corp. based on those asset values, and then

  3. determine the value of the taxpayer's ownership interest in the C Corp. with the appropriate discounts for lack of control or lack of marketability (without again reflecting any built-in tax liability).

Under this approach, step (1) would consider the tax liability, step (2) determines the value of the entity, and step (3) considers other discounts to independently determine the taxpayer's interest in the entity. An investment analyst likely would determine the investment value in that manner, with step (1) being the appropriate time/place to tax-affect the appreciated assets as of the valuation date.

In prior cases it was at step (3) that the tax liability was considered, however, and in Jelke the parties and the court

  1. determined the value of each asset (perhaps considering future growth and any tax thereon, but without considering the built-in tax liability that existed in gain as of death), then

  2. determined the value of the C Corp. based on those asset values, as reduced by any built-in tax liabilities that came with those assets as of the date of death, and then

  3. determined the value of the taxpayer's ownership interest in the C Corp.

What is not clear is the effect, if any, of applying the built-in tax liability in step (1), (2), or (3): would it make a difference in the end result?

Some critics argued that the Tax Court in Jelke committed error when it considered the tax affect in step (2), because Judge Gerber did not embrace the taxpayer's argument that future gain should be considered along with the court's evaluation of the time-value impact of delay before the tax itself might be incurred.

That is, critics argued that any discount for delay should be offset by future gain generated during that same time lag. (Those critics failed to consider the corresponding possibility that future losses might offset present or future gains.) In the estate tax context, the Jelke Tax Court was looking at a static tax liability, not one that would fluctuate in the future, perhaps because the liability for future gain (or loss) was considered in step (1) when determining the value of the underlying assets themselves, which does reflect their future gain (or loss) or income potential. In any event, none of this reflects the final result in Jelke, although it may help to explain why the court sought simplicity.

On appeal and purely for simplicity, the Jelke court rejected "prophesying as to when the assets will be sold" by instead assuming an immediate liquidation of the corporation and sale of all its assets on the date of the decedent's death (notwithstanding that the decedent's 6% interest could not command such a liquidation or sale). In valuing the business itself, and not the decedent's interest in it, the fundamental difference in the Tax Court and the appellate court approaches was this immediate liquidation assumption, embraced because it "eliminates the crystal ball [timing issue] and provides certainty and finality to valuation."

Said the court, it "also bypasses the unnecessary expenditure of judicial resources being used to wade through a myriad of divergent expert witness testimony, based on subjective conjecture, and divergent opinions. [It] has the virtue of simplicity and . . . provides a practical and theoretically sound foundation . . . ."

An odd element inheres in the court's date-of-death liquidation assumption, that a willing buyer would discount dollar-for-dollar for the full tax liability. Said the court:

[W]hy would a hypothetical willing buyer . . . not adjust his or her purchase price to reflect the entire $51 million . . . built-in capital gains tax liability? The buyer could just as easily . . . acquire an identical portfolio of . . . securities as those held . . . without any . . . underlying tax liability . . . ."

" Indeed — why would any buyer want a 6% chunk of this entity, instead of just constructing their own portfolio? Moreover, on the "forgotten" side of the willing-buyer, willing-seller equation, the Jelke dissent asks why a hypothetical willing seller would "agree to a price that ignored completely the time value of money. No rational seller would accept a price that subtracted the entire amount of the future tax liability as though it were due immediately, when that liability will almost certainly be spread out over future years . . . ."

We are again reminded that courts almost universally (but not without dissent) ignore the seller side of the equation.

As an analog (but not how the court actually performed its determination, nor how any party argued the case), one way to view the Jelke issue is to consider the gain tax liability at the date of death as if it was a recourse indebtedness that encumbered an includible asset. In such a case the accepted mechanism is to include the underlying asset at its unencumbered valuation date fair market value and then reduce that amount with a §2053(a)(4) deduction for the value of the debt.

And if that debt was not going to be incurred until some time in the future (and, unlike most debt, if the liability was not going to grow over time with an interest factor — which distinguishes the built-in capital gain tax liability from most §2053 deductible debt), then discounting to present value of the liability based on when it is likely to be incurred is an appropriate approach.

That is what the Tax Court held, using a determination that the gain in Jelke would be incurred over 15 years (based on the historic rate of asset turnover in the portfolio). That also is what the appellate court held to be error. As characterized by the dissent in Jelke:

To avoid the effort, labor, and toil that is required for a more accurate calculation of the estate tax due, the majority simply assumes a result that we all know is wrong. We can do better than that. The tax court did.

Quaere whether a court in another case would take the gain at the decedent's date of death and similarly consider it as a fixed liability as of that date, even if the taxpayer used the §2032 alternate valuation date? That might be proper with a §2053 analog, but a straight valuation adjustment presumably would discount from the same date that other steps in the valuation are performed.

Obviously there is a lot that remains to be answered in this valuation context. The most important aspect of Jelke, then, may be that the court did not consider what the tax liability might be on a sale in the future at a future value. Instead, the court considered only the value of the tax liability that existed at the moment of death (when the valuation occurs) on the gain that existed at the moment of death (which is when the liability that is being considered also is measured — only on the gain that exists at the moment of death).

The inquiry is not the tax liability on gain that may accrue in the future on postmortem appreciation. The only controversy is when the existing liability will be incurred — how much is a debt worth today, if it is not paid until some time in the future — and whether to discount for the delay. That alone is controversial enough.

Jelke is important for any entity that may not garner a basis increase for its inside assets. The dollar-for-dollar valuation discount may save more tax, sooner, than ultimately will be paid, if gain ever is recognized, making this an important issue on which we likely have not yet heard the final word.

Jeff Pennell

Steve Gorin's comments:

This case (click here for ActualText) reached the right result based on a liquidation value analysis.

The dissent has a legitimate point.  But it aimed its argument in the wrong direction.  The dissent essentially said, "Come on, people!  A rational buyer would never liquidate and pay all of the taxes at once."

That's absolutely true, but it's an argument in favor of using an earnings approach to value the company.  Instead of assuming that the company would liquidate, the court should have assumed that it would continue in existence as a holding company and distribute its earnings.

However, the corporation would have had to pay income tax on its earnings, so the earning stream would also be reduced dollar-for-dollar for the taxes that would be paid.  This would have led to similar discounts on account of the C corporation double taxation structure.

A more legitimate approach would be to consider the liquidation approach and the earnings approach, then weight them appropriately to determine how the willing-buyer, willing-seller argument would turn out.  Because of the lack of control, the way that dividends would likely dribble out, and the risk premium that would be applied in using a present value analysis, the earnings approach would be heavily discounted and likely generate even a lower value than the liquidation approach.

Then the court might conclude that a willing seller would accept only the liquidation approach and just stick with that, as the majority did.

Thus, the holding of the case is correct, but both the majority and dissent missed the chance to offer a more compelling analysis.

Steve Gorin

Paul Hood has a very different take on the case:

While this was a significant and welcome taxpayer victory, is this tax law again taking valuation principles for a ride? Is this the simplicity tail wagging the reality dog?

This result brings about a bright line test for latent capital gains tax reduction, at least in the Eleventh Circuit (and probably in the Fifth Circuit).

But, is such a bright line test any more sensible than the Tax Court's "just say no" to tax-affecting S corporation earnings?

I would say "No". Why should one assume an automatic liquidation of the appreciated assets at the enterprise level on the death of a minority shareholder? What sense does that make, other than that it is simple to compute?

My problem with the IRS appraiser's work was that he used only a 13.2% discount rate for discounting the latent capital gains tax liability for such a small stock holding. Given the small size of the stock holding, an appraiser could have easily justified a discount rate of two to three times as large as the one that the IRS appraiser used.  This would have significantly increased the reduction in valuation at the enterprise level for the latent capital gains tax.

Did the Eleventh Circuit decide this case on a matter of pure simplicity on the big issue, while leaving untouched the unconscionably low lack of control and lack of marketability discount levels that the Tax Court allowed?   I submit that, had the Eleventh Circuit properly discounted the valuation for latent capital gains tax and upped the discounts for lack of control and for lack of marketability to reasonable levels, the result would have come out pretty similarly, while making much more sense from a valuation perspective.

I would rather see the cases decided based upon solid valuation principles than by seemingly nonsensical, but easy to administer, bright line tests.  

Paul Hood


The Fifth and Eleventh Circuit Courts of Appeals now stand shoulder-to-shoulder on valuation discounts for unpaid built-in gains taxes of C corporations.

In Jelke, the Eleventh Circuit Court of Appeals allowed 100% of a C corporation's unpaid built-in tax asset as a dollar-for-dollar reduction to value as a matter of law.  The Court directly adopted, over the Commissioner's loud objections, the Fifth Circuit's methodology in Estate of Dunn v. Comm'r, 301 F.3d 339 (5th Cir. 2002).  The Court's admiration for Dunn was on full display: "… we are in accord with the simple yet logical analysis of the tax discount valuation issue set forth by the Fifth Circuit in Estate of Dunn, 301 F.3d at 350-55, providing practical certainty to tax practitioners, appraisers and financial planners alike."

Frazier Jelke III died on March 4, 1999, in Miami, Florida.  His revocable living trust held a 6.44% stock interest in Commercial Chemical Company (CCC).

CCC had operated a chemical manufacturing business from 1922 to 1974, when the business was sold and the proceeds invested in marketable securities.  CCC was still a holding company at Jelke's death.

Assets of CCC at date of death included marketable securities, valued at $178 million.  A deferred capital gains tax of $51 million lurked.  In addition, CCC held other assets valued at $10 million.

The Court recited its standards of review: – clear error for factual issues; de novo for matters of law.  The opinion finds valuation is a mix of fact and law.  Specifically, said the Court, valuation methods are a matter of law, therefore the Court reviews the methods employed by the Tax Court de novo:

"The Tax Court's findings of fact are reviewed for clear error.  Where a question of fact, such as valuation, requires legal conclusions, we review those underlying legal conclusions de novo."

The opinion contains a summary of history of discounting C Corporation asset values for deferred taxes on built-in capital gains, noting that the enactment of the Tax Reform Act of 1986 played a pivotal role.  The 1986 law imposed in all cases corporate-level taxes on distributions to shareholders upon liquidation; whereas prior law did not, under the codified "General Utilities doctrine." 

Prior to the 1986 law change, the Tax Court disallowed any such discount on the grounds that the payment of such tax was deemed by the courts to be too uncertain, remote or speculative.

The 1986 law change eventually led to admission by the Tax Court that some discount for taxes on built-in gains was appropriate, but that court never provided guidance on how to determine such a discount. 

Estate of Davis v. Comm'r, 110 T.C. 530 (1998) broke this new ground, but the discount was pronounced to be part of the lack of marketability discount. 

The Court in Jelke observed:

No liquidation of the holding company or sale of its assets was planned or contemplated on the valuation date. No tax was due and owing on the valuation date. Estate of Davis, 110 T.C. at ___. Nevertheless, citing section 5(b) of Rev. Rul. 59-60, 1959-1 C.B. at 242-43, the Tax Court determined, under an economic reality theory, that a hypothetical buyer and seller would not have agreed on that date on a stock price that took no account of the corporation's built-in capital gains tax.20 Id. at ___.

While the Tax Court in Davis opined as to how much of that discount was attributable to the capital gains tax, its method for making that determination never saw the light of day.  Eventually, it was the Circuit Courts of Appeal Cases that did so, but not early on.

in Estate of Eisenberg v. Comm'r, 155 F.3d 50, 57 (2nd Cir. 1998). 

The Court in Jelke noted a study found in the record of Eisenberg

The Second Circuit cited a (then) recent study surveying CPA valuation experts, attorneys involved in business transactions, and business brokers. The survey illustrated that a large majority of buyers of closely-held stock demanded a discount for contingent capital gains tax liability. See John Gilbert, "After the Repeal of General Utilities: Business Valuations and Contingent Income Taxes on Appreciated Assets," Montana L.Rev. 5 (Nov. 1995).

The Commissioner acquiesced in result in AOD 1999-001, 1999-4 I.R.B. 4, concluding:

We acquiesce in this opinion to the extent that it holds that there is no legal prohibition against such a discount. The applicability of such a discount, as well as its amount, will hereafter be treated as factual matters to be determined by competent expert testimony based upon the circumstances of each case and generally applicable valuation principles.

The Sixth Circuit showed it was of like mind to the Eisenberg Court in Welch v. Comm'r, (unpublished) 208 F.3d 213 (6th Cir. 2000).  But in neither case was asset value decreased by 100 percent of the unpaid built-in-gains tax.

The Jelke opinion concludes that case history leads to allowance of a discount for unpaid capital gains taxes on unrealized built-in gains as a matter of law, when the net asset valuation approach is properly used.

The Court in Jelke found Davis and its progeny uniformly rejected the notion that any showing of an imminent sale or liquidation was necessary to establish a discount for unpaid built-in-gains tax, because of the repeal of General Utilities doctrine. 

It was the Fifth Circuit in Estate of Dunn that allowed a 100% discount.

Addressing the method for determining that discount, Jelke found the present value method adopted by the Tax Court unacceptable because it was based on a showing of the term over which the tax was likely to be paid.  Requiring determination of when the tax is likely to be paid to arrive at the discount logically leaves the door open to arguing over whether the time when the tax will be paid can be ascertained at all.  That would revive the argument that the capital gains tax discount is zero because payment of the tax is too uncertain, remote or speculative – the very principle already rejected as a matter of law.  The Court said:

Cases prior to the Estate of Dunn, prophesying as to when the assets will be sold and reducing the tax liability to present value, depending upon the length of time discerned by the court over which these taxes shall be paid, require a crystal ball. The longer the time, the lower the discount. The shorter the time, the higher the discount.

The downside of this approach is that, not only is it fluidly ethereal, it requires a type of hunt-and-peck forecasting by the courts. In reality, this method could cause the Commissioner to revive his "too speculative a tax" contentions made prior to the Estate of Davis in 1998. This methodology requires us to either gaze into a crystal ball, flip a coin, or, at the very least, split the difference between the present value calculation projections of the taxpayers on the one hand, and the present value calculation projections of the Commissioner, on the other.  (Emphasis added.)

That only leaves the method of artificially assuming that all assets are sold and the capital gains tax paid on the valuation date, in this case the date of death.

It could be said the Court forgot its manners in trotting out of its closet references to crystal balls, things fluidly ethereal, hunt-and-peck forecasting, and coin tossings. It went further, complaining about burdens placed on the judiciary while further crowing about Dunn:

The Estate of Dunn dollar-for-dollar approach also bypasses the unnecessary expenditure of judicial resources being used to wade through a myriad of divergent expert witness testimony, based upon subjective conjecture, and divergent opinions. The Estate of Dunn has the virtue of simplicity and its methodology provides a practical and theoretically sound foundation as to how to address the discount issue.

Seizing upon the majority's gratuitous comments, a scathing dissent was offered up by Judge Carnes.  Essentially, while disavowing any love of tax law, Judge Carnes accuses the majority of shirking its duty of confronting complexities germane to valuation while, at the same time, employing in other branches of law the very methods that the Tax Court had used and the majority opinion rejected.  He prophesizes immeasurable collateral damage will be done to future adjudication of actions in such diverse areas of wrongful death, Federal Tort Claims Act, and bankruptcy, comparing how the Jelke reasoning would have applied in actual past opinions.

It's no secret that every deferred dollar tax has the worth of an interest-free loan.  In the case of securities investments, that loan falls due if and when appreciated securities are sold or exchanged.  Academic studies show that low turnover investing can add up to four hundred basis points to after-tax performance.  It is surprising this has escaped the notice of the appellate court in two circuits.

In arriving at the value of the unpaid tax, it is also surprising that the expert for the Commissioner used pretax returns to reach an after-tax result.  The inconsistency seems obvious. Again, academic studies exist that show what the after-tax return is on low turnover investment portfolios.

The Jelke opinion, however, correctly observes that use of a method that requires determination of the time when the tax will be paid is inconsistent with and violates the allowance of such a discount as a matter of law, since there will inevitably be arguments over whether the time when the tax will be paid can be ascertained at all.

The weakness in the dissent is in comparing this tax case to other cases.  For example, while the majority found it necessary to set aside determinations about when (if ever) taxes will be paid, there is no question in cases of wrongful death that, but for the death, a normal life expectancy can be assumed.  While some deferred taxes seem never to come due, all humans will die – and we have a robust insurance industry to tell us when that is most likely to occur.  That the Court will be bound in wrongful death cases by this tax case therefore stretches credibility.  A more careful review of the cases raised under Federal Tort Act and bankruptcy might reveal similar chinks in the dissent's armor.

Mike Jones


We certainly haven't heard the end of the debate over the proper way to value a business with high built-in capital gains.  Stay tuned to LISI for more thoughts!


Steve Leimberg


"Steve Leimberg's Estate Planning Newsletter # 1210  (November 29, 2007) at    Copyright 2007 Leimberg Information Services, Inc. (LISI).  Reproduction in Any Form or Forwarding to Any Person Prohibited – Without Express Permission.


Estate of Jelke v. Commissioner, *** F.3d ***, 2007 WL 3378539 (11th Cir. 2007), rev'g and rem'g 89 T.C.M. (CCH) 1397 (2005); IRC Sec. 2031; Estate of Dunn v. Commissioner, 301 F.3d 339 (5th Cir. 2002), rev'g 79 T.C.M. (CCH) 1337 (2000); Treas. Reg. § 20.2031-1(b); Rev. Rul. 59-60, 1959-1 C.B. 237; I.R.C. § 311(b); Estate of Smith v. Comm'r, 391 F.3d 621 (5th Cir. 2004); Estate of McCord v. Comm'r, 461 F.3d 614 (5th Cir. 2006); Estate of Blount v. Comm'r, 428 F.3d 1338 (11th Cir. 2005); Mark R. Siegel, "Recognizing Asset Value and Tax Basis Disparities to Value Closely-held Stock," 58 Baylor L.Rev. 861, 862-63 (Fall 2006). Eisenberg v. Commissioner, 74 T.C.M. 1046, 1048 (1997), rev'd, 115 F.3d 50 (2d Cir. 1998), acq., AOD 1999-01, Estate of Davis v. Commissioner, 110 T.C. 530 (1998) (citing Rev. Rul. 59-60, 1959-1 C.B. 237), and Estate of Jameson v. Commissioner, 77 T.C.M. (CCH) 1381 (1999), vac'd and rem'd on other grounds, 267 F.3d 366 (5th Cir. 2001);

All NAEPC-affiliated estate planning councils are eligible to receive a discounted subscription rate to the Leimberg LISI service. Please see more information about the offering. You may also contact your local council office / board member to find out whether they are offering the service as a member benefit.