Why Courts Dramatically Overvalue Artist Estates

When artists die, how do experts appraise the value of their work?

Georgia O'Keeffe, Jimson Weed/White Flower No. 1 (1932), Courtesy to the Artist.


This essay addresses valuing art by a single artist for estate tax purposes. The difficulty is that the art will likely be sold years into the future because the market will not absorb the bulk of the art by one artist any time soon after the date-of-death valuation date. So the issue is to determine how much art could be sold, years and years (most likely) into the future, and at what price. Conceptually, the art would be sold by the estate to an investor, or consortium of investors, for investment and ultimate resale. Its value is the amount the buyer and investor would pay to the estate for his investment in the artist’s art inventory.

● ● ●

Blockage Discounts in the Courts

Valuations of art for estate and gift tax purposes (as well as for divorce and sale) present difficult issues when the art consists of many works by only one artist, chiefly because the art market can only absorb a limited number of works, yet the appraisal must decide value as of a certain date—in an estate, the owner’s date of death. Consequently, this requires the valuation to determine the worth of a stream of income from art sales, sometimes in the rather distant future. Since it is quite clear that it would be imprudent to sell simultaneously and immediately a large number of works by the same artist, three United States Tax Court cases have accepted the concept of a “blockage discount.”

Unhappily, these three Tax Court cases involving the art in the estates of David Smith, Georgia O’Keeffe, and gifts of Alexander Calder art, while accepting the concept of “blockage” to discount the value of the art, have applied the concept in a way to greatly limit the discount, and so produce art valuations (that is, a determination of what a buyer would pay) much in excess of real-world values.

Who is the Hypothetical Buyer from the Estate: a Large Number of Retail Buyers or a Single Bulk Purchaser?

In 1972, the Tax Court first took up the issue of a “blockage discount” for the value of art for estate tax purposes.1 The artist, David Smith, died in an automobile accident on May 23, 1965, owning 425 pieces of metal sculpture created during various periods of his life, beginning in 1937.

In valuing the art for estate tax purposes, the Smith executors first computed the price of each piece if sold individually at retail at the time of Smith’s death, the total being $4,284,000. The executors then discounted this figure by 75% on the theory that these works could only be sold at the date of death to a bulk purchaser for resale, and then reduced this figure by one-third to cover the sales commissions which would be owed to Smith’s gallery, Marlborough. The resulting figure of $714,000 was reported as the date-of-death value of the 425 sculptures.

The Smith estate argued that, if all 425 sculptures were offered for sale at the date of death, the purchaser could only be a person or syndicate acquiring the bulk of the sculptures for future resale, that such a person would be required to make a large cash investment which could be recouped with an acceptable profit only over a ten-year period of time; and that such a person would only pay 25 percent of the separate one-at-time value, which, after subtracting selling Marlborough commissions of one-third, resulted in a value of $714,000.

The Smith court said that the 75% discount which the estate had applied to the one-at-time values was “too high.” On the other hand,

We think that, at the very least, each willing buyer in the retail art market would take into account the price he would be willing to pay for any given item, the fact that 424 other items were being offered for sale at the same time. The impact of such simultaneous availability of an extremely large number of items of the same general category is a significant circumstance which should be taken into account.2

The Smith Estate argued that any value must be reduced by one-third to take account of commissions to which Marlborough Gallery was contractually entitled for its selling efforts. But the Tax Court said that the date-of-death value is what the estate would receive from a hypothetical sale, that is, “what a purchaser from the estate would pay.”  (The Court clearly is thinking only of a retail sale to the ultimate buyer.) Accordingly, the Smith court held that the fair market value at the date of death of the Smith sculpture was $2,700,000, equal to a 37% discount from the hypothetical single-piece total price of $4,284,000.

The Smith Court states, what is no doubt true, namely, that a willing buyer in the retail art market “would take into account the price he would be willing to pay for any given item, the fact that 424 other items were being offered for sale at the same time . . .”  No doubt true, but not at all helpful in determining the price a bulk buyer would be willing to pay for all 425 sculptures. In effect, the Smith Court is arriving at a valuation of all the estate art by assuming a non-existent retail market, at the date of death, for each and every sculpture. The real market is a buyer or a syndicate of buyers who, in turn, would decide their purchase price from the estate by a calculation of their likely future sales (and cost of selling commissions and expenses) in both the dealer and the retail markets. Instead of trying to determine what each retail buyer would pay for a given sculpture, the Smith Court should have determined what a hypothetical estate buyer would have paid, since an estate bulk buyer would necessarily have made his own calculation for future retail (and wholesale) sales. The point becomes more clear, if one thinks of the sale by the Smith estate of the art inventory of David Smith’s “art business.” Accepted procedures for selling a business owned by an estate clearly suggest that the estate executors would not seek to sell inventory into retail market. Rather, the executors would seek out a bulk inventory buyer interested in investment and future resale in retail and wholesale markets, who would also consider his selling expenses.

O’Keeffe (Wrongly) Decides that Smith was Correct. That is, Value at Date of Death is What the Retail Buyer Would Pay

Twenty years after Smith, in 1992, the Tax Court3 again took up the issue of a blockage discount for the art of Georgia O’Keeffe in the following words:

The O’Keeffe estate employed Eugene Victor Thaw (Thaw) to appraise O’Keeffe’s works in the estate.

It was Thaw’s opinion that the appropriate blockage discount that should be allowed for O’Keeffe’s works in the estate on the date of death is 75 percent. Thaw’s opinion was based on the understanding and assumption that all of the works in the estate would be sold to a single buyer as a bulk purchase, which would require a syndicate of investors. The hypothetical buyer would have to hold the works for many years and, in determining the price to be paid, would take into account interest, selling costs, promotion, maintenance costs, and carrying charges.  Thaw also based his opinion on the assumption that a buyer would consider “fluctuations from the very high market plateau for O’Keeffe’s in 1986,” although he would have advised a potential buyer that prices of works by O’Keeffe, on average, were unlikely to go down. 4

The O’Keeffe Court noted, without elaboration, that Thaw would not have advised the O’Keeffe to sell at a 75% discount:

Thaw would not, however, have advised a hypothetical seller of the works in the estate to sell the total of those works at a 75% discount. Thaw was under the impression that determination of blockage discount required him to assume a hypothetical buyer on the date of death who would have been required to purchase all of the works of the estate in bulk on that date.  Although Thaw’s written report categorized the works in the estate by medium and price, he did not differentiate among the works in determining the discount to be applied.

In this case, the opinions of petitioner’s experts that O’Keeffe’s works in the estate, individually valued in excess of $72 million, could have been sold on the date of her death for $18 million, defies common sense.  Our conclusion is based in substantial part on the particular content of the works. The individual values of 44 pieces totaled almost one half of the agreed value of the whole group!5

The O’Keeffe court continued:

Thaw, for example, in effect assumed a forced sale in bulk of all of the works of the estate to a single buyer. Petitioner tries to justify that assumption by arguing that ascertainment of fair market value requires that property “must change hands” on the date of death. That argument is unsupported by authority or reason and ignores the concepts of willing buyers and willing sellers acting without compulsion—the defining actors in a fair market value transaction.

Both Thaw and Rose [a second expert for the estate] conceded that they would not have recommended to a seller that the works in the estate be sold for the $18 million that they opined was the blocked value of the estate on the date of death.

We are persuaded from Thaw’s testimony and other evidence that the most valuable works would not necessarily be sold first and that some works of all types and values would be fed into the marketplace at a controlled pace. The most reliable consensus of the experts is that the better works could be sold within seven years, and sale of the bulk of the estate would take more than 10 years. 6

The O’Keeffe court then introduces the concept of different markets into which certain art could be sold, again taking issue with the Thaw approach of a single bulk buyer from the estate, who, in turn, would sell into specific markets:

From the evidence, it is apparent to us that different works in the estate would be of interest to different segments of the art market. The parties have not reasonably quantified assumptions about the specific markets in which segments of the works in the estate would be salable.  Petitioner has erroneously assumed that all of the art would initially be sold to a bulk purchaser, who would purchase for resale.  Petitioner thus ignores the market of collectors who, while taking into account resale value, are not primarily interested in the rate of return on the investment.

We conclude that the respective experts in this case each failed to consider the relevant market for particular works of O’Keeffe or groups of works in the estate . . . The failure to consider the relevant market undermined the respective positions of the parties.

We are persuaded that O’Keeffe’s works in the estate on the date of her death should be segmented, not necessarily by value but by quality, uniqueness, and salability.  There should be at least two categories, i.e., works that are salable within a relatively short period of time at approximately their individual values and works that can only be marketed over a long period of years with substantial effort.  We believe that petitioner’s experts’ opinion is valid only as applied to the second category of works.  Respondent’s argument is meritorious with regard to the first category. For want of a more reliable breakdown, we conclude that one-half of the value of O’Keeffe’s works in the estate would be appropriately subject to petitioner’s experts’ analysis and should be discounted 75 percent. Using our best judgment on the entire record, we conclude that the other half of the total value of O’Keeffe’s works should be discounted 25 percent.  After considering the entire record, we conclude that the fair market value of O’Keeffe’s works in the estate at the date of death was $36,400,000.7

Thus, the O’Keeffe court decided that the O’Keeffe estate “erroneously assumed that all of the art would initially be sold to a bulk purchaser, who would purchase for resale,” and, instead, the art “should be segmented, . . . by quality, uniqueness and salability.”

But, is not this “segmentation” into various retail markets exactly the analysis a bulk estate buyer would make in deciding on a bulk purchase price? O’Keeffe decided that the estate experts on valuation “failed to consider the relevant market for particular works of O’Keeffe or groups of works . . . ” But, how else could the estate bulk buyer decide what to pay except by an analysis of the dealer and retail market for particular works, based on the qualities of each of the works?

Thus, Smith and O’Keeffe agree that in arriving at date-of-death value the court should not determine what a bulk buyer would pay for the art! Instead, both courts sought to determine (a problematic task!) the amount a single retail buyer would pay for a single piece knowing that many other pieces would be sold at the same moment.  Indeed, the assumption of a retail sale is why Smith would not subtract the cost of sale commissions, but that determination should be for the bulk buyer in calculating his purchase offer to the estate. (Of course, it is fair for the court to decide that the hypothetical bulk buyer did not properly calculate his resale opportunities, but that is very different from trying to figure out, in a bulk sale circumstance, what a hypothetical retail buyer would offer.)

The Calder Court Approaches Blockage as a Simple Determination of the
Time Value of Money

In 1985, the Tax Court8 decided the value for gift tax purposes of 1,226 gouaches created by Alexander Calder, who had died November 11, 1976. Calder’s widow, Louisa Calder, stipulated with the IRS that the average retail value of each gouache was $2,375, and reported on her gift tax return the total value of the gifts to be $949,750.

The Calder court noted the IRS “annuity” approach:

Under this approach, realization of the value of the art works can be compared to the right to receive an annuity of the stated amount over the given period, and the present worth of such annuity can be determined from the appropriate valuation tables, . . .  The appropriate valuation factor reflects a discount for the amount of time the various installments of the annuity are deferred.  As applied in the instant case, the effect is to grant a blockage discount in a somewhat more sophisticated manner than the usual method of applying a single percentage discount to the retail value of the items at the date of the gift.  We are not prepared to say that respondent’s theory is unreasonable, but its accuracy obviously depends upon the validity of respondent’s assumptions regarding the number of gouaches that can be liquidated each year and, thus, the length of time such liquidation will require.9

The Calder court went on to say that it did not have to rely on assumptions to decide how fast the market could absorb the art:

The actual sales experience in the period is the best evidence we have of the true absorption rate of the gouaches in the marketplace, and should be preferred to determine the estimated liquidation period for the purposes of determining the appropriate amount of blockage through an annuity approach. 10

Calder determined that it would take as much as 22 years to sell all the gifted art at the stipulated average price of $2,375 per gouache. Hence, the date-of-death value of the right to receive $2,375 per Calder gouache up to 22 years in the future was, in total, $1,200,000.

Calder took an annuity approach to valuation. That is, it calculated the present value of money to be received up to 22 years in the future, $2,375 per gouache.  It then simply calculated the time value of money and discounted the $2,375 based upon an assumed prevailing interest rate. However, treating the value of art as an annuity payable as much as 22 years in the future, is, of course, not in accord with the inherent uncertainty of the art market generally—all the more so for art created by one artist, even one with the reputation of Alexander Calder. Nevertheless, the principles underlying the time value of money are well understood and accepted by the courts.  The question then becomes what amount of money is to be received in the future. Calder assumed that amount of money would be the stipulated retail price, $2,375 for each gouache.

The Time Value of Money

A sale of a large amount of art by the estate of one artist requires that the art be sold gradually over a considerable period of time. For the proper estate valuation of art created by a single artist at the artist’s date of death, the assumption should be that the art will be sold in a manner that results in the net value as large as current and future market circumstances permit. That is, the date-of-death value of the estate art is the maximum amount the art would be worth to a buyer acquiring the art with the intention of reselling in a way resulting in the largest possible financial return to the estate’s buyer.  The price to this hypothetical buyer is the discounted present (date-of-death) value of a stream of revenue expected from future art sales.

All three Tax Court decisions properly took into account the time value of money in arriving at the present value of a stream of revenue to be received in the future.

The Tax Court has long recognized the common sense principle that a sum of money due to be received in the future is worth less than the same amount immediately available. How much less depends upon the prevailing general interest rates, because the time value of money is amount that could have been earned but for the delay in future sales. Thus, for example, with a 5% prevailing interest rate, $1,000 receivable in 14 years has a present value of $500, with $500 representing the discounted present value of $1,000 receivable in 14 years. That is to say, if the estate’s art could not be sold for 14 years, the date-of-death value must be reduced by 50% by reason of the time-value-of-money factor, alone.

How the Tax Court in Smith, O’Keeffe and Calder Should Have Analyzed Blockage Valuation

None of the three Tax Court decisions seem to have taken into account the single greatest risk a hypothetical buyer from the estate must have considered in deciding what to pay for the estate art, namely the risk that the art might depreciate very substantially in value over the period of time during which the art was to be sold—the so-called risk premium.

Risk and Uncertainty in Art Sales

Art to be sold in the future is, of course, subject to the usual risks of damage or destruction by fire, natural catastrophe, theft, accident, and deterioration.

But art created by a single artist (especially) is subject to another kind of risk, namely the art can decline (or increase) in value relative to any general trend in art prices. Art purchased cheaply may become very valuable ten years later or may collapse in market value over the same period.  The result of this depreciation or appreciation—the possibility of bi-directional uncertainly—is not a “wash.”

The distinguished economist, Professor William J. Baumol, gives several arresting examples of this concept:11

Just consider a government that promises to pay the holder $1,000 later today, and a lottery ticket that will pay the holder after the drawing this afternoon either $1,500 or $500 with a 50-50 chance of either outcome.

Clearly the treasury bond will be worth approximately $1,000 because it is virtually cash in hand and will actually be so in a few hours.  The lottery ticket . . . brings with it the possibility that anyone who purchases it for $1,000 will be $500 poorer that evening.  Consequently, the market will normally deduct what is described as a risk premium for the price of the lottery ticket.  Normally, moreover, the magnitude of this risk premium will be greater the greater the range of uncertainty entailed in the investment. Thus, consider a second lottery ticket that offers either a $2,000 payoff or one equal to zero . . . the lure of the investment brings the possibility of doubling one’s money if one pays $1,000 for the ticket.  But, now, the danger is that one’s investment will be lost altogether . . . The second lottery ticket, because it entails the greater gamble, can be expected to entail the larger risk premium . . . 12

Professor Baumol continues:

The very fact that there patently is a range of possibilities . . . means that the . . . future value of those art works must be reduced by a substantial risk premium to obtain a correct current [date-of-death] valuation.

. . . the big risk, is the risk that the price of the work will drop unexpectedly and disastrously in the future.

. . . it has happened to Titian and Sargent, among many others.

. . . the very great risk of investment in art is confirmed by every statistical investigation of which I am aware.

. . . The starving artist who goes unrecognized during his lifetime but whose works are later sought after avidly by museums and wealthy collectors is the stuff of romantic story-telling.  We all know of van Gogh who never was able to sell a picture while he lived (except for one sale to his brother), of Gauguin and Cézanne, whose paintings commanded no high prices and whose works have since undergone price explosions. There is, however, an abundance of cases in which histories are reversed . . . 13

Historical Winners and Losers

The economists Bruno S. Frey and Werner Pommerehne14 wrote in 1988 about their important historical study of art prices and investment return:

Man in Black by Frans Hals sold at Christie’s in 1885 for five pounds sterling and in 1913 for 9,000 pounds at Sotheby’s, and artists like Manet, Matisse have shown enormous price increases. But John Singer Sargent’s oil sketch San Virgilio sold for 7,350 pounds in 1925 but in 1952 sold for only 105 pounds at Christie’s. Other well-known artists such as Sir Edwin Landseer and Sir Joshua Reynolds had works sell at real financial loss.

The Unpredictability of Fashions: Resulting in Unanticipated Price Movements.

Frey and Pommerehne continue:

A careful analysis of the long-term evolution of the price of paintings suggests that the changes in price can be understood as a random process. The long-run rate of return on investments in paintings cannot be attributed to a systematic identifiable factors . . . The price movements that in retrospect are termed “fashions” are not predictable even by art experts, and therefore cannot be anticipated or and exploited systematically.

. . . There are many examples that support these findings:  Vermeer was held in low regard for a long time, and his paintings accordingly fetched low prices; today his paintings are among the world’s most expensive. The same can be said of El Greco.  Turner’s paintings at first sold at very high prices; then prices fell sharply. Today his prices fetch record prices . . . . Conversely, there are many artists who were previously highly esteemed, but nowadays are little known . . . Alma-Tadema also comes to mind; his works were initially highly priced, but by the time of his death in 1913, the prices of his paintings had been falling for a decade . . . one of his most famous paintings could not find a buyer at 5% percent of the price paid in 1904. 15

Additional Risks and Uncertainties: Tending to Depress the Value of Art

If the art valuation, as it often is, is in the context of estate values, of course the deceased artist will not be adding to the market supply with the sale of newly created works. But it is important to bear in mind that, in calculating the rate at which the market can absorb art from the estate, owners other than the artist’s estate will be selling into the public and private markets, possibly at a greater rate than when the artist was alive.

Also, art created by the artist remaining in the estate at the artist’s death is, by definition, art that was not sold during the artist’s lifetime, most likely because it is less saleable than that which the artist sold during life. (Clyfford Still may be the exception, proving the rule).

Conversely, if the valuation is for purposes of divorce or sale where the artist remains alive and creating art, this market supply overhang would tend to reduce the rate at which the market would be able to absorb the artist’s existing inventory.

The net result of the above-noted risk and uncertainties affecting the value of art, and, most importantly, the risk premium, suggests that many single-artist estates are very much overvalued by both appraisers and courts.


New York, New York
January 2016

Ronald D. Spencer, Esq.
Carter Ledyard & Milburn LLP


1 Estate of David Smith v. Commissioner, 57 T.C. 650 (1972), aff’d, 510 F.2d 479 (2d Cir. 1975).

2  Smith at 658.

3 Estate of Georgia T. O’Keeffe, 63 T.C.M. (CCH) 1992-210 (1992).

4 Id.

5 Id.

6 Id.

7 Id.

8 Calder v. C.I.R., 85 T.C. 713 (1985).

9 Calder, at 724.

10 Calder, at 725.

11 Affidavit of William J. Baumol, In the Matter of the Estate of Andy Warhol, Index No. 1987-0824 (Surr. Ct. N.Y. County).

12 Id.

13 Id.

14 Bruno S. Frey and Werner W. Pommerehne, Is Art Such a Good Investment?, The Public Interest, Spring 1988, at 79.

15 Id.


Editor’s Note

This is Volume 6, Issue No. 1 of Spencer’s Art Law Journal.  This winter issue contains two essays, which will become available on artnet in February 2016.

The first essay discusses problematic valuations of single artist collections which often result in substantially overvaluing the art.

The second essay examines legislation currently proposed to amend the New York Arts and Cultural Affairs Law, with the goal of protecting art experts and thereby encouraging these experts to express opinion about the authenticity of visual art.

Three times a year, this journal addresses legal issues of practical significance for institutions, collectors, scholars, dealers, and the general art-minded public.



Follow Artnet News on Facebook:

Want to stay ahead of the art world? Subscribe to our newsletter to get the breaking news, eye-opening interviews, and incisive critical takes that drive the conversation forward.
  • Access the data behind the headlines with the artnet Price Database.
Article topics