Market Sentiment Indices for Fair Market Value
SOURCES OF INFORMATION
AMR’s data is compiled by in-house analysts from auction results provided by auctioneers.
The price a picture reaches at auction can be influenced by many factors: condition, subject matter and size are just a few. We recommend additional research when buying and selling fine art, such as consulting reference books and obtaining advice from experienced sources.
Media covered by AMR’s Create an Index Service:
- Painting (Unique Two-Dimensional Works of Art) – includes Paintings and Works on Paper [Paintings – oil, tempera and acrylic. Works on Paper – watercolours and drawings, including collages, gouache, pen, pastel, crayon, pencil, charcoal, ink and chalk.]
- 3D – sculpture and 3-dimensional works, including installations.
- Prints – etchings, lithographs, engravings, lino and others.
The price recorded by AMR is the ‘hammer price’, which is called out at auction and at which the lot is ‘knocked down’ to the bidder.
Many Auctioneers levy an additional charge to the buyer. This is known as the ‘buyer’s premium’ and is not included in the price recorded by AMR.
FAIR MARKET VALUE
A term referring to an object’s likely sale price if offered by a willing seller to a willing buyer. Since the auction process brings buyers and sellers together to compete in an open market setting, a sale at auction is considered to be a measure of Fair Market Value.
From the sale prices of the works of the artists in AMR’s database, it is possible to indicate a typical level of their performance at auction in recent years. AMR considers that the best way to do this is show the average of the central 80% of prices. AMR also publishes the number of prices (Lots) used in calculating each index and endeavours to publish only those indexes in which the number of observations is measurably consistent from year to year.
NOMINAL / REAL TERMS
AMR offers subscribers the option adjust data for inflation using the Consumer Price Index in the USA and the Retail Price Index in the UK.
AUTHENTICITY AND ATTRIBUTION
AMR cannot account for the authenticity and attribution of the works recorded. The integrity and reputation of the auctioneer does this. In recording the name of the artist, it is the policy of AMR generally to accept the style of presentation used by the auctioneer in the sale catalogue. If there is no clear authorship, such as ‘attributed to…’, ‘circle of…’ or ‘studio of…’ etc. then AMR does not record the sale.
1. PAINTINGS (unique two-dimensional works of art), SCULPTURE, PHOTOGRAPHS and PRINTS
1.1 AMR uses price data from auction sales worldwide – capturing a significant proportion of sales by value – to construct its price indices. In addition, AMR adjusts price data to filter out extreme values and seasonality, both of which are liable to exert distorting effects on any constructed price index.
1.2 Prices are recorded in pounds sterling (or the sterling equivalent in the case of a foreign currency sale) on the day of the sale.
1.3 For cases where the auction house includes a buyer’s premium in the published price, AMR removes this premium to ensure the recorded price used in the production of indices properly reflects the value of the item only.
1.4 AMR produces data points by taking average values over 12 consecutive months.
1.5 Extreme values (some of the largest and smallest values) are removed using a simple trimming process which captures the central 80% of sales by value.*
1.6 The final price index is computed as the ratio of the current period moving average to that of the base period, multiplied by 1000 to set the base period index value equal to 1000.
1.7 AMR applies a smoothing calculation by averaging each 12-month period with up to five preceding 12-month averages.
1.8 Historic inflation and exchange rates are recorded allowing indices to be inflation- adjusted and/or produced in any major currency as required.
1.9 Indices are produced for both individual artists and classes of art**.
1.10 ‘New’ artists are those achieving index status after 1975 for which there is no 1975 base period. For these, it is necessary to calculate a pseudo-base period average by extrapolating backwards from their first recorded price data points. An average is calculated for the first two years of available data for each new artist. These values are then related to a corresponding group to which the artist belongs or, failing this, to the Art 100 index. Indices for the new artist are then calculated from the pseudo-base period as though it were a true mean price.
* Different price brackets can be selected to create indices as required. Examples include 100% of values, the central 80% and the central 50% of ordered values, as well as percentiles such as the top 2% and the top 10%, also the top and bottom quartiles.
** Indices for classes of art can be calculated using ‘group’ or ‘pool’ methods before the simple moving average is applied. ‘Groups’ calculate means for individual artist data separately and ‘Pools’ pool all artists price data together.
2.1 AMRD’s antiques and collectibles indices are based on a number of items that are used to represent the ‘class’. The index for a class represents a comparison of the average market value for items in the class relative to their value in the base period, 1975 (or some later date).
2.2 The classes fall into several groups, as shown by the Chinese Ceramics example below.
Chinese Ceramics: Chinese Ceramics Han to Yuan 20 items, Chinese Ceramics Ming to Qing 20 items, Chinese Ceramics Export Porcelain (1550-1850) 20 items Chinese Ceramics – General 60 items
2.3 The database consists of valuations (in sterling) of each one of the items at three- monthly, six-monthly or twelve-monthly intervals from 1975 (or some later date) to the present. All the valuations in the base period (1975) were actual prices paid for these items in 1975 (or in certain cases estimates of values by experts).
2.4 AMRD then produces price indices for antiques and collectibles in the same way as for paintings, sculptures, photographs and prints, as outlined in 1.1 – 1.8.
3. ‘ALL ART’ Index
3.1 AMR uses price data from auction sales worldwide – capturing a significant share of sales by value – to construct its All Art indices.
3.2 For each of the artists in AMR’s database, an ‘Artist Price’ is calculated by averaging sales over a 24-month period.
3.3 Weights are attached to observations in the 24-month window with proportionally more weight placed on more recent sales.* (Anticipating a degree of market sentiment may affect prices, the length of time which has passed is taken into account.)
3.4 A mean value is calculated from all ‘Artist Prices’.
3.5 Processes 2.2 to 2.4 are repeated monthly.
3.5 The entire history of the series is smoothed with factors which decrease exponentially.
3.6 Artists rankings are measured by weighting the Artist Price according to frequency of the artists’ individual sales. Weightings are therefore produced by calculating the number of lots sold for each artist as fraction of the total number of lots for all artists in the index.
* For infrequently traded artists, the absence of new sales from one month to the next would mean no new price could be determined until a new lot was sold. Where no monthly sales occur, the most recently calculated price remains “live” and unchanged until a new sale is recorded.