Rubles per Square Metre: Why Market Prices Do Not Measure Artistic Value

A market price is a dated fact about a transaction. It records what one buyer paid one seller, in one currency, under one set of laws, on one day. It is strong evidence about the buyer. It is weak evidence about the paint.

The ledgers of my field prove it. Between 1874 and 2008, Russian and Soviet paintings moved through four pricing systems: a collector’s chequebook in the 1870s, a state export programme in 1929, cooperative purchase tables in the 1950s, and an international auction room after 1988. The same primed canvas could carry numbers three orders of magnitude apart across those systems. The canvas had not changed. The institution around it had.

I paint on the standard numbered canvas formats — No. 20 at 73 by 60 centimetres, No. 30 at 92 by 73, No. 40 at 100 by 81 — on linen sized with hide glue and primed with a chalk ground I cook myself. I buy those materials by the metre and the kilogram, and I can check them against matter. A sale price is the one number in a painting’s file that cannot be checked against matter. This article is about why.

The State Priced Canvas by Format Number

The Art Fund of the USSR purchased work from union members at fixed rates. Its purchase tables priced a canvas by format number and by the artist’s union category: a No. 20 sat below a No. 40 on the same page, and the payment rose with the primed square metre. The standard support was cotton duck on a pine stretcher; the standard grounds were chalk in hide glue or a factory half-oil; the standard white in Soviet tube paint was, by the 1960s, titanium dioxide. Those are the material facts the tables were written against.

When a buyer pays by size, the rational artist paints larger. Late-Soviet exhibition halls filled with No. 50 canvases (116 by 89 centimetres) and No. 60 canvases (130 by 97) partly because the categories paid best at that scale. You can read the price list off the stretcher bars in any provincial museum store: measure the stored stock, and the formats cluster where the tables clustered.

The unofficial artists of the 1960s and 1970s sold outside those tables, in apartments, at prices set by conversation rather than category. Their supports record the exclusion. Ilya Kabakov’s Ten Characters albums, begun in 1972, are ink, pencil and watercolour on Soviet paper — a material with no format number and therefore no table price. Erik Bulatov’s earliest paintings are on board, not canvas. A market-free position looked like something in matter: cheap, light, and foldable.

Tretyakov Paid 92,000 Rubles for the Turkestan Series

In 1874 Pavel Tretyakov bought Vasily Vereshchagin’s Turkestan series for 92,000 rubles. A Moscow professor then earned about 3,000 rubles a year, so the purchase equalled roughly thirty professorial salaries. The Apotheosis of War (1871), oil on canvas, 127 by 210 centimetres, hung inside that purchase and hangs in the Tretyakov Gallery today.

The figure measured a project, not a rank of genius. Tretyakov was assembling a national gallery — he handed the collection to the city of Moscow in 1892 — and he paid a nation-building price for a series built to hang as one public statement. In the same years he paid a few hundred rubles for single genre canvases from the Peredvizhniki exhibitions. Both numbers are documents of what one collector believed a national school needed on its walls, and at what ratio. No auction existed in 1874 to arbitrate the ratio. The 92,000 rubles are a budget line, not a verdict.

The Hermitage Sold Raphael to Fund a Plan

In 1928 the Soviet government decided to sell Old Master paintings from the Hermitage for hard currency. In 1931 Andrew Mellon paid 1,166,400 dollars for Raphael’s Alba Madonna, a round panel about 94 centimetres across, and 502,875 dollars for van Eyck’s Annunciation. Both paintings hang today in the National Gallery of Art in Washington, which opened in 1941 on the collection Mellon assembled.

The dollar figures are documents of the First Five-Year Plan, adopted in 1928, which needed convertible currency for imported machine tools. The state had decided that a Raphael could be converted into industrial equipment, and the price recorded the exchange rate it set for that conversion. Nothing in the panel’s own layers — oil over gesso on poplar — changed between 1930 and 1932. The number moved because the Plan moved.

This is the cleanest case in my field of a price measuring an institution and nothing else. Raphael did not improve in 1931. The 1,166,400 dollars measure one government’s need for foreign currency, stated to the cent.

Sotheby’s Brought Hard Currency to Moscow in July 1988

In July 1988 Sotheby’s held the first Western auction on Soviet soil, in Moscow, with the state taking a share of the proceeds. Grisha Bruskin’s Fundamental Lexicon sold for about 240,000 dollars. At the official rate of 0.6 rubles to the dollar, that was 144,000 rubles. The average Soviet monthly wage in 1988 was about 200 rubles. One triptych therefore equalled roughly sixty years of an average salary.

Between 1987 and 1988 the paint on those canvases did not change; the room did. The same artists had sold through Art Fund tables and commission shops for three-figure ruble sums. The 1988 figures measure the first contact between a closed domestic system and buyers holding convertible currency: a scarcity premium, an exit premium, a novelty premium. Scarcity is a fact about supply. It is not a fact about quality.

The material record shows the border crossing. Works painted in apartments on cotton duck and board in the 1970s were, by the mid-1990s, being painted on imported Belgian linen over pine stretchers, because fee income in hard currency could buy it. In my own studio the change came in 1992, when the first roll of Belgian linen replaced state cotton duck. The format numbers stayed the same — No. 30, 92 by 73 — while the cloth under the chalk changed.

Sixty Million Dollars for 88.5 by 71 Centimetres

In May 2008 Kazimir Malevich‘s Suprematist Composition of 1916 sold at Sotheby’s in New York for 60,002,500 dollars, then the highest auction price recorded for a work of Russian art. The canvas is oil on linen, 88.5 by 71 centimetres. Nothing in its material constitution explains the figure. The number was made of law.

The chain is checkable. Malevich left the canvas in Germany in 1927. The Nazi authorities confiscated it as degenerate art in 1937. After 1945 it entered the Stedelijk Museum in Amsterdam. In 2008 the city of Amsterdam settled the heirs’ long-running ownership claim, title was cleared, and the painting was consigned to New York. Remove the settlement and there is no sale, because clouded title is worth a fraction of clear title at auction. Add the Russian export law of 15 April 1993 — most works over one hundred years old cannot leave the country, and works aged fifty to one hundred need Ministry of Culture permission — and the pool of 1916 suprematist canvases that can legally reach an international buyer shrinks to a handful. The 60 million measures clear title, legal mobility and scarcity. It measures the paint hardly at all.

A conservator’s report on that canvas — ground composition, craquelure, retouching history — would tell a historian more about 1916 than the sale price can. The 2008 number tells us about 2008: two bidders, one settlement, one law. The oil and the ground were the same before the gavel and after it.

What a Price Is Good Evidence Of

Read as institutional documents, prices are among the best evidence an art historian has. The Art Fund tables tell me the state wanted large, legible canvases and paid by the primed square metre. Tretyakov’s 92,000 rubles tell me what a nationalist collector in 1874 believed a public gallery needed. The Hermitage ledger tells me what a machine tool cost, priced in Raphaels, in 1931. The 2008 Malevich figure tells me what clear legal title is worth in a scarce category. Each number is dated, named and checkable — as evidence about demand.

Read as verdicts on merit, the same numbers are noise. A No. 30 canvas of 1916 and a No. 30 canvas of 1918 can differ by a factor of thousands for reasons of law alone. A triptych painted in a Moscow apartment could not fetch 240,000 dollars in 1987 and could in 1988, with the same pigments from the same tubes. An instrument that sensitive to the room and that insensitive to the paint cannot be measuring the paint.

The whites make the point in matter. Lead white can blacken in sulfur-laden air; zinc white cracks and delaminates on a flexible cotton support; titanium white keeps its colour but no oil film survives a century unaltered. A price cannot tell you which white sits under the surface, and the difference between the three will decide what the painting looks like in 2100. I keep a cost book for my own work — metres of linen, grams of chalk, litres of hide glue, hours of priming — and that ledger is the only price schedule I can check against matter, square metre by square metre. It tells me what a painting costs to make. It says nothing about what a painting is. The market’s ledger runs the same direction, only larger: it tells you what a painting cost its buyer’s project, not what the painting is.

FAQ

Is an expensive painting always a better painting?

No. The same canvas carried a ruble price and a dollar price within a single year in 1988, with no change in its materials. The Hermitage’s Raphael fetched 1,166,400 dollars in 1931 because a Five-Year Plan needed currency, not because the panel improved. Quality is argued from the object — its ground, its format, its pigment layer — never from the invoice.

Did Soviet artists really sell work by canvas size?

Yes. The Art Fund of the USSR purchased from printed tables scaled by format number and union category, and the standard formats — No. 20 at 73 by 60 centimetres, No. 30 at 92 by 73, No. 40 at 100 by 81 — were the units in which those prices were written. Measure a late-Soviet museum store and the formats cluster where the categories paid best.

Why did the Malevich reach 60 million dollars in 2008?

Three institutional conditions. The 2008 settlement with the city of Amsterdam cleared title for the heirs. The canvas stood outside Russia, so the export law of 15 April 1993 did not block the sale. And 1916 suprematist canvases with exportable legal status are scarce. Bidders paid for clear title and legal mobility. The oil and the ground were the same before and after the gavel.

What does the 1993 export law do to art prices?

The law of 15 April 1993 requires Ministry of Culture permission for works aged fifty to one hundred years and bars most works over one hundred years old from export. A price in London or New York therefore includes the legal cost of exit. Two comparable canvases, one painted in 1968 and one in 1918, differ in price partly because one can still leave Russia with paperwork and the other cannot leave at all.

How should a historian use auction prices?

As dated institutional documents, cross-checked against the object. Establish who bought, under which law, in which currency, on which date. Then examine the support, the format number, the ground and the pigment layer. The price explains the transaction. The materials explain the painting.

What This Column Does Next

Next month I open the Art Fund tables themselves: the format numbers, the category rates, and what per-metre pricing did to the look of late-Soviet wall painting, from exhibition halls to the monumental commissions that paid by the running square metre of decorative panel. If you hold a purchase receipt, a commission-shop tag, or a 1988 catalogue with prices marked, write to me. A price is a document: date it, name the parties, name the law it was made under — and then put it down and look at the paint. The paint has been answering longer, and it does not inflate.