![Sample our Museum and Heritage Studies journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5945/TOC-Subject-Sample-2021-Museum-Heritage-Studies.jpg"})
Abstract
The financial crisis of the past several years has taken a visible toll on the arts landscape. Financial crises, moreover, have been growing more frequent and more severe. To combat this trend, the arts must explore options to help them finance their operations more self-sufficiently. A financial instrument that has become relatively important to the art market in recent years may provide an opportunity for art museums to develop a new strategy to supplement their current funding sources. The instrument, known as an ‘auction guarantee,’ provides a risk optimal method for art museums to generate revenue to supplement their endowment, membership, and admissions income. The auction guarantee is introduced and discussed, and some possible methods for museums to raise the funds necessary to exploit the potential of the auction guarantee are considered.
Acknowledgments
Thank you to the many individuals who took their time to read, comment on, and speak with me about this paper. Special thanks are due to Nic Iljine, Brian Kennedy, David Kusin, and András Szántó. Thank you to Stephanie Hare at Oxford Analytica for allowing me to develop some of the ideas expressed here and to Heinrich Schweizer for critiquing some of my more grandiose proposals. Thanks also to the anonymous reviewers whose incisive comments helped me develop and reframe my argument. My deepest gratitude is due to Gregg Rubin and Katherine Jentleson for their collaboration over the past few years. This paper is better for all of the help I have received; any remaining errors are entirely my own.
Notes on contributor
Adam Levine is the 2013 Mellon Foundation Fellow at the Toledo Museum of Art. He received his D.Phil. in the History of Art at the University of Oxford and was the co-founder of Art Research Technologies.
Notes
1. It goes without saying that there are many types of museums of which art museums are a single variety. For the sake of convenience, however, the term ‘museums’ will be used to refer to art museums exclusively. Moreover, the model offered here is applicable to an even smaller subset of institutions: large-scale American (art) museums.
2. Another reason to diversify revenue sources is that the predominant source of earned income for larger institutions, their endowment, may face problems meeting expectations in the future. Recent work suggests that the ex-ante equity risk premium (ERP) may be only 4%. Put less technically, if we generously assume an average Treasury bond yields 2.5%, an endowment should expect on average a 6.5% annualized return, which falls short of the 7–8% returns assumed by many museums in their budgets. On the ERP, see Asness (Citation2011).
3. The percentage discount is calculated relative to the average of the low and the high estimates provided by the auction house before the sale at which the object was bought-in.
4. A third, nonsale option is for an owner to take out a loan against a work of art (an increasingly prevalent practice).
5. The asking price for a work can be, and often is, adjusted on the private market. The incentive for the middleman, who takes a percentage of the sale price, is to offer a high initial price. As a result, when prices are revised they tend to be revised downward (otherwise the lot would sell at the initial price). A private market sale by consignment, in other words, is equivalent to a descending auction. Although no work has been done on the private art market, following Vickrey, descending and ascending art auctions are functionally equivalent and the discounts for an unsold work should converge (or at least approximate one another). See especially Vickrey (Citation1961) and also Ashenfelter (Citation1989). For the effect of (negotiated) guarantees on auction house revenue, see Greenleaf et al. (Citation2002).
6. Theoretically, the dealer could buy works that s/he believes will appreciate in value over time. The reasons to question the long-term viability of this solution are too many to catalog here (cf. Baumol Citation1986), but the widespread failure of ‘buy-and-hold’ art funds underscores the problems with the model (cf. Horowitz Citation2011, Appendix C).
7. A guarantor could also pay the difference between the high bid and the guarantee price. See note 8 for an example. Note the similarity to the model proposed by Greenleaf, Rao, and Sinha (Citation1993).
8. It is conceivable that a guarantor could allow the high bidder to retain ownership of the artwork for US$700,000 and pay the seller the US$50,000 difference. This model has operational complexities that are beyond the scope of this paper.
9. At auction there are two sale prices – the ‘hammer price,’ which is the price of the winning bid, and the ‘price plus buyer's premium,’ which is the price paid by the buyer inclusive of auction house fees. Since the seller normally does not receive any of the auction house's fees, the guarantee is applied to the hammer price.
10. Sothebys had US$18.7 million in outstanding guarantees at the end of their 2011 fiscal year; no mention was made of the total issuance. Note also that the increase in guarantees from 2009 to 2010 indicates demand for guarantees. The market for guarantees evaporated due to a lack of supply, not a destruction of demand (see ‘Assessing the Risks of a Guarantee-Driven Acquisition Strategy’ section).
11. For a similar argument for loan guarantees, see Merton (Citation1977). For the not unrelated proposal of the application of Credit Default Swaps to art lending (and by extension to guarantees), see Campbell and Wiehenkamp (Citation2008).
12. Museums, of course, still do acquire exceedingly valuable works of art as gifts, through purchases endowed by donor funds, etc. Suffice it to say that the relationship between museums, acquisitions, and donors (acting as the intermediaries that help to achieve the acquisitions) is highly schematized in the current discussion. The primary point for the argument here, however, is that museums have trouble acquiring objects using only their own monies in today's market.
13. On the heavy-tail of art prices, see Campbell (Citation2004) and Bocart and Hafner (Citation2012).
14. The capital employed for guarantees could conceivably be used in multiple auction events each year. The returns for each auction event should be more or less consistent, which is to say that an assumption of a net 20% return is, in fact, only an assumption of four auctions with net 5% returns (or three with net 6.67% returns, etc.). This, of course, also assumes all of the available capital is put to work each auction cycle.
15. It is assumed that any additional surplus – that is, the difference between the deficit and the guarantee returns – would have been reinvested into the acquisitions budget. It also is assumed that revenue from guarantees would not be taxed as unrelated business income.
16. Of this US$50 million, only US$20 million is unrestricted (the market value of unrestricted endowment and nonendowment assets, by contrast, is more than US$836 million).
17. Proceeds from a deaccessioned work are used only to acquire other works of art – the proceeds are never used as operating funds, to build a general endowment, or for any other expenses’ (Association of Art Museum Directors Citation2007, 1).
18. Censure by peer institutions could prove especially difficult in the event further actions are taken against an offending museum (i.e., membership to museum bodies could be revoked, loans could be withheld, etc.). This is to say that while the ethical issues are not black-and-white, they are nevertheless sufficiently divisive to be the crux of the matter for many.
19. Note that this model does not motivate museums to simply sell off their entire collection since, in the event a museum ‘loses’ all of its money on unprofitable guarantees, the result would be that the museum would have replaced any artwork sold with new artwork of an equivalent (monetary) value. A museum's cupboards will never run bare since the ‘option’ is backed by real property.
20. In auction-theoretic terms, the artwork has a private value to the museum divorced from the market value of the artwork, so there is no ‘winner's curse’ per se. See Goetzmann and Spiegel (Citation1995).
21. It is of note, however, that recently even the American Association of Museums (now the American Alliance of Museums) has called into question whether or not future museums ought to retain their tax exempt status (cf. American Association of Museums Citation2012, 8).
22. To service the demands of these clients, the auction houses have developed another instrument known as the ‘irrevocable bid,’ which, for the purposes of this paper, can be thought of as functionally equivalent to the guarantee. In point of fact, the auction house still underwrites the guarantee itself and then ‘sells’ an irrevocable bid to a third party to cover some or all of the auction house's exposure.
23. These regulations – meant both to enable museums to underwrite guarantees and to discourage pure speculation backed by the value of a permanent collection – could provide the beginning of a framework to allow museums to operate in the guarantee space without forfeiting their tax exempt status.