Thursday, September 11, 2014

$800 Million or $8 Billion for Detroit's Art? It Depends on Who Is Asking

The current dustup over the collection of the Detroit Institute of Arts, which involves many contentious issues in the political and legal arenas, is fascinating from the perspective of fine art appraisals.  The bankruptcy case provides front-row seats as some of the top appraisers in the country publicly assign values to the same works of art.  While each is doing their best to develop credible results, they are reaching vastly different conclusions. The appraisal profession struggles mightily with the concept of subjectivity, but as the Detroit case illustrates, it is difficult to reach a consensus when valuing works of art in the hypothetical.  It also demonstrates that no appraisal exists in a vacuum: the viewpoint of the client as well as any other intended user of an appraisal naturally comes into play during the process of developing an opinion of value.

The breaking news that the various creditors may be settling probably renders all of the work that the appraisers have done in this case moot.  However, I think the case shines a bright light onto something with which the appraisal profession is a bit uneasy: appraisers adopt a point of view which will benefit their client.  However USPAP clearly states that an appraiser “must not advocate the cause or interest of any party or issue”.  So what gives?  My view is that what matters is that the decisions and methods that an appraiser uses must be credible.  When USPAP speaks of advocacy, it means that an appraiser can’t cherry-pick comparables, or ignore facts that affect value; every close comparable must be dealt with and all relevant facts need to be considered.  The document must be credible.  But we aren’t dealing with stocks or bonds here; works of art are unique and the market for them opaque.  Reasonable people can differ over almost every aspect of an appraisal.  So naturally, as an appraiser goes about their work, they are going to adopt the point of view of the person who hired them.  The report has to be defensible, reasoned, and proper in every way.  But if you are not taking your client’s point of view in a subjective situation, you aren’t doing your job.

Jan Steen's Gamblers Quarreling, c. 1665
COD's appraiser puts this at $1 Million and DIA has it insured for $1.5 Million;
The two opposing appraisers assign values of $10 Million and $13.5 Million

All of the below appraisals were taking place while the so-called “Grand Bargain” was being struck, whereby a pile of money, from private foundations as well as the State of Michigan, would be paid to creditors in exchange for taking any of DIA’s art off the table, forever.  They are still arguing over the legality of this transaction, as well as over who should get whatever money does actually get transferred, but, most importantly from the appraisal angle, the various parties are sparring over how large the pile of money should be.  In other words, irrespective of whether or not any of DIA’s collection is sold, it now matters how much it is worth.  The methodology used and the type of value established are where the rubber meets the road.


In August of 2013, the emergency manager of the City of Detroit announced that Christie’s was being retained to evaluate the part of the DIA collection comprised of works acquired with City of Detroit (COD) funds, explicitly excluding any work that was donated. They were also asked to give advice on how to monetize the collection without selling it. The odd assignment resulted in an odd document, produced in December of last year, that provides values for the works and then goes on to explain how not to sell them, which is probably not a  typical assignment for an auction house.  Christie’s cover letter states that Fair Market Value (FMV) was the valuation they arrived at and pointedly claims that these values were not auction estimates, but then they go on to give wide ranges of values which look exactly like what you’d see in an auction catalog!  Their report also does not give any comparables, although they claim to be using the market data approach.  And perhaps worst of all, neither the appraisal nor the cover letter makes any reference to USPAP.  At this point in time, one would expect a major institution like Christie’s to be adhering to the professional guidelines while performing appraisals.  But I digress.

The appraisal covers almost 2,800 objects; however they eliminated almost 1,000 as being of little or no value.  The remaining 1,741 objects were then valued at between $454 and $867 million (you may see the report here  the very important cover letter which details the parameters of the appraisal is here).  The appraisal itself is divided into three parts:  works on view, works in storage with a value above $50,000 and then stored work valued below $50,000.   As might be expected, most of the value was concentrated in the first group, the 326 works actually on display in the museum galleries. 

This assignment is no joke; they had to appraise 1,741 objects in about 4 months, knowing that their values would immediately be made public and that more than a few people would look at them as greedy lowlifes, aiding and abetting the unholy process of stripping DIA of its art and angling to get the works on consignment for themselves.  Christie’s sorted through almost 2,800 objects from the entire history of humanity, spanning very diverse regions and cultures.  By focusing on the works actually on display, they cleverly narrowed the focus onto the works which are the most valuable.  Furthermore, they presumably did not have to bother the DIA art handlers, and were able to simply walk around the museum.  So, right off the bat, instead of eating up time appraising 2,800 objects, they looked closely at 326 and, by their reckoning, captured about 90% of the total value of the collection.   Then, in going through the various lists from DIA, they found another 80 works with a value of over $50,000, which they itemized, and the rest they simply lumped together in a single number.  Along the way, Christie’s could not resist patting themselves on the back for reattributing two works of Asian art which they found to be much more important than DIA had realized.

Christie’s is a major force in the art world of course, with resources, manpower and a huge network at their disposal to rely on when providing valuation services.  So, despite the lack of comparables in the report, one has to lend considerable credence to their efforts.  I do think it is odd to simply zero out 1,000 works.  They are described as a collection of photographs of little interest, a group of paintings by local artists and some coins and so on.  But really, zero?  I mean, not even $100?  I suppose this is a rounding error in the whole scheme of things, but these are the sorts of decisions that along the way begin to add up.

More importantly, the values that Christie’s comes up with are puzzling.  Their cover letter says that they are establishing FMV, with all the ‘willing buyer/willing seller/relevant marketplace’ boilerplate. However, they go on to say that the buyer’s premium has not been considered, which is always included when figuring Fair Market Value, at least in the USPAP and IRS definition of the term.  The wide range for the works is also strange, with the final top number almost double the low one.  They explain that the lower number is the more conservative and the higher the “most advantageous” between the willing parties.  As both Elizabeth von Habsburg and Victor Wiener (appraisers who later evaluated DIA;’s collection for some of COD’s creditors) point out in their depositions, having a range is sometimes OK in many appraisal situations, but a range this wide is borderline useless in most case.  In this instance, however, it serves their client well. 

The tricky nature of the assignment is perhaps one reason that Christie’s was being so fuzzy about the type of value they came up with as well as the big range.  Their client, the Emergency Manager of Detroit is trying to save the city.  He wants Detroit to become a functioning municipality with a future that has a hope of meeting its obligations, with a community of taxpayers who stay put and as such wants to have a world-class art collection within the city limits. He doesn’t want to sell it. But he also needs to be seen as doing his best to pay the City’s debts—it is his responsibility as fiduciary, so he commissioned the appraisal and is here going through the motions of ‘putting every asset on the table’.  At the same time, he has lined up a group of donors to protect the collection who he can’t really afford to fleece as he needs them for a bunch of other gifts down the road.  So the appraisal is rather vague on a number of important issues such as what exactly the art is worth and by what definition the value is being calculated.  Such flexibility allows the Manager and COD’s lawyers to use it in a variety of ways to achieve the Grand Bargain and to make it look like the best possible outcome. 

The rest of Christie’s report proposed various ideas for creating a revenue stream from the collection without selling it, such as creating a “Masterpiece Trust” which would be accessed by a consortium of museums who would pay DIA for the privilege.  As it doesn’t really concern the appraisal of the works, I will not discuss these ideas here, except to again note that the report is a real hybrid behind which the overriding goal almost certainly is to deflect the option of selling off the collection.

Diego Rivera's Power of Industry fresco from 1935
While all the other appraisers assumed this could not be safely removed from the wall,
Victor Wiener Associates argues that it can and thus has significant value


During the wait for this document, a group of creditors (the ones most likely to receive the lowest amount for their claims) petitioned the judge to appoint a panel of independent appraisers to do the job instead of Christie’s, but the judge turned them down, despite the obvious benefit to the employment prospects of independent appraisers. (Nicholas O’Donnell has an excellent collection of posts about the legal aspects of the case here). However, there was much more to come.  Blocked from assembling an appraisal panel, the aforementioned group of creditors solicited bids from prospective buyers of parts of the collection as well as from a lender willing to make a loan to Detroit taking the collections as collateral.  In many ways this makes a lot of sense:  what better way to determine the value of anything than to actually get a bid? 

The offers are worth a look. (The Art Market Monitor covered  this matter quite comprehensively here).  According to court documents, 4 entities indicated interest in the collection:

-Poly International Auction house offered ‘up to$1 Billion’ for all the Chinese art, subject to their examining it.

-Yuan Capital, an investment group with partners offered between $895 Million and $1.4 Billion for 114 works of their choice.

-Bell Capital Partners, another investment group offered $1.75 Billion for everything (not sure if that included the empty vitrines, but it probably did).

-Art Capital Group offered a $2 Billion loan against the entire collection, with an interest rate of between 6 – 9%

Like an appraisal, these numbers are hypothetical.  The prospective participants did have to provide some proof that they had the ability to close any proposed transaction and that they weren’t simply shills for the creditors, but they could easily back out of any deal and, of course, the works aren’t actually for sale at this point.  So, like the various appraisals, these offers become less about reality and more about jostling for position in the legal cage-match.


In July, 2014, the COD revealed that it had commissioned an appraisal of the entire collection from Artvest, an ‘independent advisory firm that provides investment advice for the art market”, according to their website.  This is not an appraisal company, per se, but again the COD was not looking for a straightforward appraisal.  As was the case with Christie’s, the COD directed Artvest to not only come up with values, but to come up with some ideas of how to generate income from the collection without selling it.  Artvest was also tasked with exploring how feasible it was to actually sell the whole collection, as well as to examine the proposed deals that the creditors had generated. 

The legal strategy here is easy to discern:  by putting the whole collection forward, the COD again was ostensibly following the emergency manager’s claim that ‘everything is on the table’.  The creditors had obviously made enough headway with their demands that the entire collection was up for grabs that the COD had to address it.  The COD was also stealing a march on any creditor who was appraising the collection as a whole and taking the opportunity to diminish any possible impact that the offers for parts of the collection might have.

Artvest comes up with a number between $2.7 and $4.6 Billion, and then lands on the midpoint of about $3.68 Billion. They arrived at the number in the following manner:  First, they simply accepted Christie’s appraisal of 2,800 COD works, taking the mid-point of their high/low range.  This makes sense; Christie’s is obviously a reputable company and probably the mid-point was what Christie’s was intending to be used all along.  Besides, they still had a lot of objects to appraise. 

Artvest then undertook their own appraisal of remaining works in the collection that they identified as high value, by which they meant anything listed on DIA’s insurance as $1 Million or higher, and/or which they spotted during their own walk-through of the collection which they thought would be over $750,000.  To do this work, they hired four appraisers [two of whom I know professionally:  Betty Krulick is the President of the AAA, of which I am a student member and aspire to become a full-fledged member, and Sabine Wilson, who was an instructor of mine at NYU].  This step also seems sensible; they were trying to spend their time wisely, focusing only on the high-value items, and a filter of the insurance values combined with a walk-thru seems credible to me.

The work the appraisers did is credible as well.  All four were striving to establish FMV and in most cases, they actually list the comparable that they felt was the closest, which made for very pleasurable reading, at least for me.  Added up, the four appraisers assigned values of between $1.6 and $2.4 Billion, which Artvest averaged to $2.05 Billion.

However, this covered only another 423 works, meaning they still had another large group to account for.  Their next step was to basically group everything into various categories (Contemporary, American Furniture, Islamic, and so on).  They then took the average value for objects under $750,000 in Christie’s report, using the lower estimate and multiplied that by the number of objects in each category, and neatly appraised about 57,000 objects.  In this equation, they also excluded anything that was listed under $5,000 (I assume on DIA’s insurance).

This last step I find a bit off.  I certainly recognize that it is not possible to appraise over 60,000 without a team of 300 appraisers working for a year.  (By the time a smaller team finished, the market would have moved!)  So obviously, some assumptions have to be made.  But for these 57,000 objects, they came up with a value of $971 Million.  Bear in mind, they were creaming off anything above $750,000 as they walked around the museum.  That certainly leaves a lot of money on the table, as I assume there must be a number of $500,000 works in the collection.  Interestingly, Artvest makes the argument that this methodology actually overstates the value, since a lot of what is in a museum collection is low-end stuff that they simply agreed to accept as gifts in order to get a few choice items from wealthy donors; in other words, to flatter collectors they agree to accept an entire collection when only a few things are museum quality.  This is debatable, but deft, giving cover to a rather sweeping decision.  In my opinion, to take 57,000 objects in a major museum and pencil in $971 Million just doesn’t pass the smell test.

I also question the decision to zero out everything under $5,000.  Artvest states that it would cost more to catalogue, ship and sell these works than $5,000 in their reasoning, but I disagree. Christie’s zeroed out 1,000 objects; it is likely that Artvest has zeroed out ten times that.  They are dismissed along the way as pottery shards, arrow heads and textile fragments.  Surely these things have a value. I mean, mightn’t someone out there like to own an arrow head from DIA and be willing to pay, say $20?  Or $30?  And maybe there are a few really good ones.   Artvest argues that it would cost more to sell them than their value, which again might well be true.  But this is something one can control for later in the appraisal by, perhaps applying some discounts and so on in a later step.

And in fact, Artvest does one whole heck of a lot of adjusting.  Indeed, although they used very credible appraisers to establish FMV, Artvest refers to the all of the values as “indicative value” which is not an appraisal term as far as I am aware.  An internet search reveals it to be an investment term which is a preliminary estimate of the bid/ask of a given security.  And this is exactly the manner in which Artvest uses the numbers.  Having established what they hope is a credible ‘indicative value’ Artvest proceeds to apply all sorts of discounts and scenarios to the top line and end up with a range of values from about $1.84 Billion to as low as $850 Million dollars, for the whole shebang. 

Bear in mind that evening sales of contemporary art are considered a failure if they fail to crack $400 Million in one night.  How on earth can the entire contents of one of the best museums in the United States ever be listed at $850 Million?   There were at least two offers for more than that for parts of the collection!

Taking a look at the report, it makes the following points about the market in general, relying heavily on their interpretation of Clare McAndrew’s latest TEFAF reports:

-People think the art market is hot, but it is only the big ticket items that sell well; the middle and low part of the market is soft.  While maybe 500 objects in the museum are masterpieces, a lot of it would fall into the ‘soft’ part of the market.

-Three of the four hottest categories in the market have declined since 2011.

-Except for Contemporary, the art market has reached a plateau and the new money from the BRIC countries that fueled the current boom is waning, so the rest of the market is probably going to go down in the next five years.

-Contemporary art has peaked and it too could reach a ‘breaking point” and go down soon.  Lots of savvy collectors are selling, which is another sign of a top.  And when it all crashes, it crashes hard, like in 2008, when everything went down by 50%.
-Because it is a bankruptcy situation, the DIA would not be able to sell privately and would have to go through auction, which is more risky.  And the two best auction houses wouldn’t agree to handle most of the small-ticket items in the collection, meaning smaller firms would handle the sale and the prices would be lower and the buy-in rates higher.

Reading through this report, you would have thought that the art market had been in a slump over the past few years, except for the occasional masterpiece that comes up at a contemporary sale, and that it was about to cease functioning altogether in the near future. 

What am I missing here?  Is it possible that the art market is so bad?  Is everything I see and hear showing that a lot of people are buying a lot of art a mirage?  I do agree that selling art is not as easy as it looks and that we needn’t get carried away by all the glamour and hype, but I feel Artvest goes overboard.

Then, turning to DIA’s situation in particular, Artvest notes:

-A lot of what DIA has is out of favor at the moment (like American Art in particular)

-A DIA sale would be boycotted by a lot of collectors, museums in particular would refuse to buy and even if they didn’t, they don’t have any money to buy art anyway, so a private sale to another institution would be off the table.

-The big ticket items would attract a lot of attention at a sale and the notoriety would work against achieving high prices.

-Selling everything at once means you need to look at liquidation value, as well as apply a blockage discount.

-Things would inevitably be bought-in, which would further reduce their value, as they would be burned.

-The amount of work on offer would be way too high for Christie’s or Sotheby’s to offer guarantees, increasing risk.

-Since there is so much work, it would be best to sell things over time, which also increases risk, so if you don’t apply a liquidation and or blockage discount, you have to apply a discount that takes into account the time factor.

-Litigation would prolong and hinder the sales process at every juncture.

-Controversial museum sales in the recent past have not done well at auction.

Some of these are good points, some seem a stretch.  A later report by Victor Wiener rebuts a lot of these claims so I won’t go into it much here.  What is important to note is that Artvest is saying that the DIA provenance would hurt rather than help a given work’s prospects at a sale.

The above factors lead Artvest to apply various discounts in different scenarios.  In every case, there is at least a 50% discount applied to the value before reaching the bottom line.

These scenarios are:

A.      An immediate sale to a single buyer (requiring a discount to get a liquidation value or at least blockage)

B.      Applying all the other possible discounts besides blockage/liquidation just because the whole situation is screwed up and also because the Contemporary market is about to crater

C.      Assuming a 3-5 year window to sell everything, requiring a discount for the risk of the passage of time and for things buying in

D.      A 3-5 window, but where litigation drags things out.

Then, for each of the above, Artvest runs the numbers twice, once at the mid-point and once at the low estimate of their ranges—never the high.

In short, the Artvest report takes a very dim view on the entire matter; it opines that at most, the creditors would be lucky to get $1.8 Billion.  Again, all of these assumptions are ones that an appraiser is allowed to make.  And the decision to go low is, of course, dictated by the client.  The question is, is this credible to start low and then discount heavily?

Really, the Artvest document cannot really be called an appraisal at all; they themselves do not refer to it as such. It is properly an “Expert Witness Report” and as such it is really a partisan legal filing meant to buttress the argument that the Grand Bargain is fair.   Again, the point of view of the client has affected the final numbers.

Renoir's Woman in Armchair from 1874;
Insured by DIA for $14 Million and appraised by Artvest at $10 Million;
Winston assigns a value of $15 Million and VWA appraised this at $22.5 Million


While the COD was working on their side of the story, two creditors took matters into their own hands and hired their own appraisers.  Winston Art Group’s Elizabeth von Habsburg was hired by Syncora and Victor Weiner Associates was hired by FGIC to develop their own theories of the monetary value of DIA’s art.  Both of these clients are bond insurers who stand to get hammered in the bankruptcy; if they can sell off the art to get paid in full, they would, and if they can get the Grand Bargain to kick in more money, that also helps them. (They were also involved in generating the offers mentioned above).  So their motivation is to get a higher number than Christie’s and Artvest.  [I note here that I have had professional interactions with Winston Art Group and Ms. von Habsburg is on the board of the American Appraisers Association; Victor Weiner has been my instructor for two appraisal classes at NYU].  The Wiener appraisal can be found in its entirety on the State of Michigan website; I have been unable to find a copy of the Winston report, but a lot of the information in it appears on Wiener’s and her deposition can be found online.

The Wiener appraisal, along with his long, contentious deposition, makes for a great read.  Unlike Christie’s or Artvest, VWA puts a single specific number on the collection: $8,149,232,354, although it immediately states that the number is ‘probably more than that’.  The value given is Marketable Cash Value (MCV), which is what it sounds like:  the amount of money a seller would get after all commissions and expenses have been taken out of the selling price.  VWA and Artvest are the only two appraisals that look at the collection as a whole so I will focus on them together here.

It is interesting to consider what might be the most appropriate value in this situation.  When things are being donated for charitable purposes, or when they are gifted under non-charitable circumstances, or when an estate is being settled, the value used is FMV.  One might argue that under the Grand Bargain, the work is not being sold, and is, in a sense, being donated or gifted, so FMV makes sense.

However, VWA’s client wanted them to consider the possibility of using the DIA art as collateral, so in that case MCV is much more appropriate.  Furthermore, even though it is a lower number, it allows the creditors to say that this is what one could actually get.  And course, even though MCV is typically lower than FMV, the appraisers came up with a much higher number than Artvest.

What was their methodology?  I should start by saying that finally in all of this appraising, VWA mentions adhering to USPAP, which is a great relief!  Disappointingly, most of their comparables are contained in the workfile and are not publish, but it should be noted that VWA is claiming the high ground by invoking USPAP.

VWA’s first step was to appraise 387 objects themselves.  It is somewhat unclear exactly how these were chosen, as they range in value from $175 Million down to $2,000.  In his deposition, Wiener states that they were trying to at once capture the really expensive items, but to also create a sample, appraising objects from all the various departments at DIA, which would then help them later on in their appraisal.  Like appraisers before them, VWA gives a range, in this case $3.09 – $4.04 Billion and lands on $3.56 Billion for the first batch.  Since these are all individually appraised, using comps, one would need to actually see the comps to critique them, but the methodology as described is sound.

Second, VWA took all of the 596 works that they themselves had not appraised but that had been individually appraised by either Christie’s/Artvest as well as Winston, and then averaged the midpoints of their valuations and came up with $311 Million.    Again, this is a decision an appraiser has to make to get a project done under the deadline but it must be credible.  Is it credible to take an average of other values and simply use them?  It is certainly problematic.  For one thing, all the other values purport to be FMV.  VWA is reporting MCV, so it’s not apples to apples, at least on the face of it.  In his deposition, Wiener stated that because they had done their own appraising of a lot of things and compared them to the other appraisers, they felt everyone else was too low, so by taking their FMV, they felt it was ‘implicit’ that these were really the lower Marketable Cash Value.  Also, VWA spends considerable time attacking the numbers of both Christie’s and Artvest, so it does make it a bit harder for them to simply take their word for it, but again, by averaging Winston in with it, VWA felt as if they had controlled for a lot of these errors.

Then, like everyone before them, VWA had almost 60,000 objects left to deal with. 

In the third step, they took everything that was left over that was given an insurance value by DIA and not appraised by Christie’s/Artvest or Winston, just over 16,000 objects .  Here it gets very interesting.  VWA recognizing that insurance values (let’s call them Retail Replacement Values, although DIA doesn’t) are unreliable.  Firstly, they might be out of date and secondly RRV is almost always way, way higher than MCV.  So, to control for this they calculated the average age of the listed RRVs and they also compared the RRVs to works that they themselves had appraised.  This gave them a factor with which to adjust DIA’s RRV, and they gave the group a value of almost $759 Million.

Here again, the appraiser is struggling to find a way to credibly appraise a huge amount of disparate material in a short amount of time.  Relying on RRVs requires a few leaps of faith, and VWA does acknowledge that in their report.    Values used for insurance do not always correspond to the realities of the marketplace and the manner in which they are generated is not always the most rigorous.  Nonetheless, these values have to come from somewhere and DIA, being a world-class institution would presumably spend some time making sure that their collection was properly insured, especially when objects were being moved, conserved or loaned.  Furthermore, VWA compared a sample of objects it had appraised on its own with the insurance values that DIA had listed and made a reasoned effort to extrapolate based on these results.  Because it relies upon individual values assigned to individual pieces by someone connected to DIA, I do find this to be credible, as unorthodox as it might be. 

VWA then had just under 43,000 works of art left after these three steps.  This large group of items had not been appraised by anybody and had no insurance values assigned by DIA.  What they did was use the groupings that DIA used to describe the works (Asian, Islamic, Old Master, Contemporary and so on).  They then went to the 2013 auction sales results from Christie’s and Sotheby’s and figured out the average selling price for each work within each category.  They then decided to add a premium or subtract a discount based on how strong that particular market is.  VWA refers to this as a “pricing matrix”

So, for example, there were 1,679 objects on the DIA inventory listed as “Ancient” in the final group.  VWA calculated that the average selling price at the two major auction houses in this category in 2013 was $80,049.  So they multiplied 1,679 to 80,049.  Then, because they think that the ancient market is strong and because the DIA provenance would enhance the desirability of the works as well as insure that the pieces would not violate international laws governing the trade in antiquities, they add a multiple of 25% for this category.  So, 1,679 x 80,049 = $ 134,402,271.  Then, the 25% premium brings it up to $1,680,02,838.  (VWA has a slightly different number in their report, page 214, likely due to rounding). 

They do this for every category and come up with about $3.5 Billion for that group.  I have thought about this a lot and read the VWA report a number of times and I don’t think the pricing matrix approach as they use it here works.  For one thing, I tried to follow along with the data they used (they relied on pages 67-69 of the Artvest report which tabulates the number of lots and values at the two houses in 2013) and could not replicate the factors they were using…I come up with lower average prices than they do in every instance.  But more importantly, I feel that their method overstates values.  For one thing, they are only looking at what works sold; they are not adjusting for the fact that some things buy-in.  In other words, they are not dividing the gross sales numbers with the total number of works offered for sale but rather only those that actually sold.  Furthermore, as COD’s lawyers point out, VWA is taking the top two auction houses and looking at their sales which include some very important items and using those results to appraise a selection of works which has already been filtered a number of times for high-value items.  Indeed, the works they are using this formula on have NO INSURANCE VALUES listed.  I am willing to imagine that DIA has a few pricey items in there.  But a lot of it must be things that would never appear at any auction, let alone Christie’s or Sotheby’s. 

VWA’s final number is about $8.1 Billion, almost $3.5 Billion higher than Artvest’s. Furthermore, their $8.1 Billion is MCV, whereas Artvest’s is an ‘indicative value’, subject to at least a 50% discount.  Clearly, the two sides are very far apart.

VWA’s report is laden with arguments for why they think the DIA collection is worth more than what Artvest says.  Their report attacks a great deal of Artvest’s assumptions and conclusions while arguing that instead of discounting works because they come from DIA, one should be augmenting them.

Specifically, VWA states:

-Michael Plummer, who signed the Artvest report is not an appraise, the report is not USPAP compliant.

-The DIA collection is incredibly important, assembled through the efforts of some of the richest and knowledgeable collectors of the 20th Century.

-A museum provenance enhances works at auction and many collectors see deaccesioning as a great opportunity to acquire great works.  (VWA here deftly quote their adversary’s experts, Christie’s, who say as much in a number of their own publications.)

-Celebrity adds excitement and a DIA sale would be the “Sale of the Century” resulting in a feeding frenzy.

-The current market pays a premium for works of extraordinary quality, of which DIA’s collection contains many.

-Artvest misinterprets and cherry picks data from the TEFAF report, which, in VWA’s opinion shows that auction numbers are down because of a lack of supply of quality works, which would not be the case with a DIA sale.  Artvest also uses 2011 to show that a peak has been reached, but overall, the market for art has not peaked when looking at a longer timeframe.

-The new money flowing into the market from emerging economies has only begun and is not about to dry up.
-The steep  rise of the art market in the past 10 years is not ‘once in a lifetime’ as it happened before in the late 1980s.

-By using limited time frames, Artvest incorrectly shows that a number of collecting categories are out of favor.

To make a number of these arguments, VWA hired two outside consultants to prepare reports, Zhang Yi, who was a co-author of the recent TEFAF report as well as Dr. Janet Barth, an economist with a specialty in the art market.

Overall, they do a pretty good job of pounding on Artvest.  As I indicated above, I feel that Artvest went way too far in the grim picture they paint of the art market.  On the other hand, it may well be that VWA is being too aggressive.  For example, while a DIA provenance would certainly appeal to a lot of collectors, when work is being sold that is of major importance it will already have an astronomical price.  Would the fact that a masterpiece that is unobtainable except in extraordinary circumstances really be sold for more just because it is from DIA?  Wouldn’t everyone bidding on Van Gogh’s Self Portrait already know what their maximum bid would be irrespective of the fact that it came from DIA instead of, say, a private collection?  But in general, VWA dismantles Artvest’s rather shaky arguments, and hiring outside experts to assist in this adds greatly to the credibility in this regard.


Winston Art Group, like VWA, is a major player in the appraisal field.  As mentioned above, Elizabeth von Habsburg, a principle at Winston, headed the team that appraised part of DIA’s collection.  Unfortunately, I have not been able to obtain a copy of their actual report, although information from it is included by VWA in their report, and von Habsburg’s deposition regarding it are available online. In total, they examined 582 objects from throughout the collection and assigned a FMV of $1.74 Billion.  Unlike Christie’s, they use the accepted definition of FMV, which includes the buyer’s premium and/or any other commissions due the seller. 

It is not clear how the 582 works were chosen; von Habsburg stated that it was her clients who gave her the list to work off of.   Looking at the range of values they provided as well as the material, it seems to me as if it was trying to capture all the masterpieces and then do a sampling of the rest of the collection.  My guess here is that the Winston report is meant to do what VWA’s does: illustrate that the COD is coming in too low, but, unlike VWA,  without relying on any statistical gymnastics to try and figure out what the whole thing is worth.  If they can take the top 100 works, show that Artvest and/or Christies are too low on a case-by-case basis, they will have scored some points for their client.  Also, by doing a sample, they can give their side more ammunition to attack the findings of Artvest and Christie’s.

The works Winston appraised to not synch exactly with all the others, so it is hard to quickly say if they are higher or lower.  Without doing a full-scale analysis of their values, it seems as if they are sometimes the lowest and sometimes the highest.  In general, though, it seems safe to say that they come in quite a bit lower than VWA and a little bit higher than Artvest.   None of their values seemed way off, at least in the areas that I am familiar with, and it seems reasonable to assume that they could defend their valuations.

What is interesting to think about with Winston and VWA is that, presumably, the two creditors who commissioned these appraisals coordinated their efforts with each other.  I have no way of knowing this, but it would stand to reason that since they have the common goal of undermining the COD’s appraisal of the DIA collection, they might check in with each other as they go about this task.  By having VWA do the whole collection and Winston do a sample, the strategy may have been to give both of their lawyers ammunition in the fight.  However, the two existence of two reports also allowed the COD to play them off one another.


Late in August, the COD deposed both Victor Wiener and Elizabeth von Habsburg and then later in the month petitioned the judge to exclude VWA’s report.  As is typical in these legal filings, the language and the tone was withering.   It begins by stating that “Victor Wiener is an appraiser who purported to appraise the entire 60,000-plus collection of art at the Detroit Institute of Arts (“DIA”) in less than
two weeks—a feat that even Mr. Wiener admits had never been achieved in the history of art appraisal.”   It goes on to state that his approach is a ‘mishmash method’, which is ‘slapdash’ and unreliable. 

COD’s major points in this are:

-A lot of VWA’s values are too high…’off the charts’

-Step 2, where other appraisals are simply averaged is unorthodox

-The Step 2 averaging uses numbers that VWA criticizes in other parts of its report

-Step 2 uses other appraiser’s FMV numbers but VWA is calculating MCV

-Step 3 uses values from DIA; these insurance values were old, unreliable and anyway, using insurance values to calculate MCV is a no-no

-VWA admitted to making mistakes which they corrected, but there are likely other ones

-The supplements that VWA applies in Step 3 are arbitrary, cannot be independently tested and the Step 3 methodology has never been used before and in fact was invented by VWA and used for the first time in the appraisal

-Step 4, where the remaining works were compared to average prices at Christie’s and Sotheby’s inflates the values because the works sold at auction are much better than the works in this group, which had been picked of most of the good stuff

-And again, Step 4 is a new invention

COD buttress their argument by pointing out a couple of instances where VWA and Winston were
miles apart on a couple of Rembrandt prints and they quote from von Habsburg’s deposition where she stated that usually an appraiser should never rely on unverified values on spreadsheets or use insurance values to calculate FMV.  (She was not referring to VWA in these statements but rather was responding to a general question).

COD makes some good points.  As I stated above, I think Step 4 is suspect and it is also true that VWA makes it hard to follow along with them in their decision-making. On the other hand, a lot of what Artvest could be criticized in the same way.  They worked quickly, invented new ways of analyzing the data and propose discounts which are impossible to test.  There really is not much difference in the leaps that each firm took in preparing their report, at least in the abstract sense.  If one wants to say that their methods are wrong, one has to deal with them.  Of course some new methods have to be invented here, no major museum’s collection has ever been used as a bargaining chip in a bankruptcy proceeding before.  

In fact, the entire proceeding is an exercise in wild conjecture: how much income will a toll road generate 23 years from now?  What amount of tax revenue can Detroit expect to generate in 2024?  What return will the retiree’s fund earn next year?  In every instance, the answer depends on your point of view.  So when Christie's considers 1,000 objects to be worthless and then Artvest does the same with 10,000, or Victor Wiener Associates adjusts insurance values to calculate marketable cash value, they are not doing anything different from the rest of the experts who are involved in the case and they are making an honest effort to be of use to the court...and their clients.