Sunday, 8 July 2018

Contrasting Rubens portraits at London auctions

Image result for rubens clara serena
Sotheby's and Christie's each offered Rubens portraits at their auctions last week. The Christie's picture, Clara Serena Rubens (above) had been recently deaccessioned by the Metropolitan Museum of Art in New York, and sold by Sotheby's in 2013 as the work of a follower. Back then it made $626,500 against an estimate of $20k-$30k. It's been promoted hard as a Rubens, including spells on loan to the Rubenshuis and the National Gallery of Scotland in Edinburgh. A dealer described the estimate of £3m-£5m as low, and predicted that she would 'break records'. It didn't sell.

I saw it at the Rubenshuis, exhibited alongside certain works by Rubens, and didn't think it stood up. It's a really endearing image and beautiful in parts, but the ambiguously sculptural treatment of the bust seems wrong for an intimate portrait of his daughter, and the application of paint seemed quite different from the other Rubens in the exhibition I saw. I'm not a Rubens specialist, and I know that some thought it was right even after seeing the Rubenshuis show. But a lot of Rubens specialists weren't quoted in the catalogue and haven't committed a view either way. Scholars are cautious of criticising new attributions, which are often uncertain rather than plainly wrong. But I am uneasy about museums showing controversial works with full attributions just before they're sold.

The Rubens at Sotheby's, which I wrote about in my preview post, was unquestioned and made £5,416,500 against an estimate of £3m-£4m. The market gave its verdict on the relative merits of the two portraits. The wonderful portrait attributed to Dürer - my favourite in the Sotheby's sale - made £1,150,000, over a cautious £300k-£400k estimate. It's still good value for something so rare and so good. Last time I can recall a plausible Dürer offered at auction was the Portrait of Michael Wohlgemut from the Schäfer collection at Sotheby's in December 1992, where it was unsold against an estimate then of £600k-£800k.
 Image result for sotheby's liberale verona
Sotheby's had a lot of pictures to my taste. I've always liked the mysterious Jacobus Vrel, and I was pleased to see one of his best pictures sell for £838k (estimated £300k-£400k). Cassone panels are often disappointing because they get so damaged, kicked and scrubbed over the centuries. The Liberale da Verona Triumph of Chastity was one of the most beautiful and well-preserved to appear on the market, and deservedly made £1,330,000 (estimate £400k-£600k, detail above). I was more surprised to see four South Netherlandish panels from the early fifteenth century sell strongly for £2.65m (est £1m-£1.5m). They're enormously rare, but I thought not of the finest artistic quality. I preferred a really fine Adoration of the Magi by a follower of Van der Goes, which made just £250k (est £200k-£300k).

At Christie's an attractive Gerard David Holy Family made £4,846,250 (est £1.5m-£2.5m) and a major portrait by Ludovico Carracci that emerged in 2005 made £5,071,250 (est £3.5m-£5m).

In drawings, Turner's spectacular Lake of Lucerne made £2,050,000 at Sotheby's. I thought it would do better. That's less than half the Blue Rigi, which is a steep discount for a less sexy title. Van Goyen drawings are common, but an especially beautiful one sold for £68,750 against a £20k-£30k estimate. There's a strong market for the very best drawings. The very best drawing sold last week, in my view, was the incredible early Fuseli of The Faerie Queen that made £728,750 (est £150k-£250k), which is both a high price and a bargain.
Image result for tuscan madonna christie's thirteenth century

There were a couple of high prices in the day sales. At Sotheby's an attractive and commercial Netscher made £274k against £60k-£80k estimate. At Christie's a thirteenth century Tuscan Madonna and Child enthroned with angels (pictured) sold for £992,750 against a £30k-£50k estimate.

Wednesday, 4 July 2018

Auction bargains

The glitzy evening sales are a precious opportunity to see masterpieces that might not return to public view for a long time. There's usually a handful of optimistic attributions and bad-but-commercial pictures, too. The day sales are more of a mixed bag, but prices start really low and quality rises really quite high. I know that objectively £5k or £10k is a lot of money, but art has always been expensive, and collectors at every price point make sacrifices. Put off replacing the car (or cycle instead), choose cheaper holidays, plan to work a little longer, or just sell the kids. A lot of these pictures are within reach for many people.
Image result for sotheby's swabian prophet
I like the early Netherlandish and early Germany pictures. Some of the studios turned out high quality pictures in some quantity. These Old Testament prophets from the Swabian school are great! Just £40k-£60k for the pair at Sotheby's. They have the same estimate on a beautiful Annunciation from the circle of the Master of the Housebook, just a foot and a half high. Look for pictures from the 'Antwerp School' for bargains. There are a few conventional names given to identifiable groups of pictures, but a lot of pictures from Antwerp studios are unidetified but often high quality. This St Jerome is particularly nice, with an excellent landscape, estimated at just £20k-£30k at Sotheby's. A little later is a large Martyrdom of St Sebastian (£25k-£35k at Christie's) by Cornelis van Haarlem, an artist of mixed quality, but rather under-rated in my view.
Image result for christies circle mor
Christie's says this is from the circle of Anthonis Mor. It's not the only attribution I disagree with, but wherever it's from this is a fine portrait (£10k-£15k). It stands up well displayed just off the main gallery at Christie's, among the evening sale highlights. Another anonymous picture is this fine Roman school Head of a Man, which will be worth much more than £7k-£10k if someone can identify the artist.
Image result for christies rubens lion

Rubens's prolific studio produced a lot of good pictures. There's a huge price change between 'Rubens' and 'studio', so a big incentive to over-attribute. On the other hand, pictures given to the studio can be cheap. Christie's has a version of the Washington Daniel in the Lions' Den without Daniel, by a follower of Rubens (£25k-£35k). And let's face it, it's the lions we really love! Big kitty cats; Internet, do your thing.
Image result for christies giolfino
Do have a look at the catalogues; they're big sales with lots of interesting pictures. At the lower end, this Nicolò Giolfino St Roch is quite charming at £10k-£15k at Christie's, and Sotheby's has a Jean-Baptiste-Marie Huët portrait for £7k-£10k. A beautiful Ceruti still life of chestnuts is estimated at £40k-£60k at Sotheby's. I like it more than a lot of evening sale still lifes.

Monday, 2 July 2018

Summer auction highlights

Image result for dresden mars
It's bronze-o-rama this month. Really great bronzes rarely come up for sale, but this month there are three in London. My favourite is by the greatest sculptor of small bronzes, Giambologna. The Dresden Mars (£3m-£5m, above) is a heart-stopping masterpiece. Exquisitely detailed and with beautiful patina, it's also a consummate summary of mannerism with its subtle and not-so-subtle distortions. The massive foreshortened hand is marvellously expressive with detailed veins and an exaggerated radius bone that looks almost like a ganglion cyst. The calf muscles are over-sized, and more distortions become evident when you study it. But the effect is artistic rather than awkward.

Sculptures are hard to display. They need protection from curious hands, but they can't really be appreciated in vitrines. Sculptures are often seen as the poor relations of paintings, and don't get the same curatorial attention. That's why auction viewings are so worthwhile. The auctioneers do a much better job of showing their wares, and you can really appreciate the quality of this masterpiece. It's a logical acquisition for the Getty, which has developed a choice collection of sculptures. I hope they get it, because they display their collection so well.

Christie's also leads with bronzes. There's a great group of Hercules overcoming Achelous by Tacca, an artist in Giambologna's studio. A gilt version of this is in the Wallace Collection, and comparing the two really emphasises the quality of the Christie's bronze. Estimate is 'on request', circa £5m. They also have a magnificent rediscovered masterpiece by Giradon, a large bronze of Louis XIV on Horseback (£7m-£10m).
Image result for durer man green background sotheby's
The picture that grabbed my attention was this outstanding Portrait of a Man against a green background, plausibly attributed to Dürer. I don't know if it's by him or one of his close followers like Schongauer, but whoever it's by, it is a masterpiece. Condition is clearly compromised; the background looks horrible. It might have been overpainted and then cleaned. But the face itself is well-preserved and fabulous quality. This kind of picture is rare outside Germany and the estimate of £300k-£400k is modest, reflecting its small size and the diminished impact from damage to the background. The excellent catalogue entry gives more background on disputes over its attribution, which is welcome. Continuing with the northern Renaissance, Sotheby's also has a rare picture by one of my favourite artists, Hans Baldung. The Holy Family with Five Angels (£2.5m-£3.5m) is rather worn in the key parts, but other elements are still finely preserved. And they almost never appear for sale. Hugo van der Goes is another rare and prized master. The Adoration of the Magi at Sotheby's is only by a follower, but I like it a lot. And over 2m wide, it's unusually large and is good value at the estimate of £200k-£300k.
 Image result for rubens venetian man sotheby's
Speaking of attribution disputes, Rubens was enormously prolific and pictures by him and his studio often appear at auction and there's sometimes a fine line between the master and his school. The best this time is a fine Portrait of a Venetian Nobleman at Sotheby's (£3m-£4m), which looks even better in the flesh, with an ambiguous sense of swagger and vulnerability. Christie's has a newly-attributed portrait with a slightly higher estimate (£3m-£5m), which I don't love. Christie's also has a fine large studio version of a lion's den, derived from the Washington Daniel in the Lion's Den, without the lions (£25k-£35k).
Image result for christies triburtine
At Christie's there's a fine large Zanobi Strozzi Last Judgment which represents Fra Angelico's beautiful style (£2-£4m) and a superb early Spanish masterpiece by Miguel Ximénez, also of the Last Judgment (£600k-£800k). My favourite is a small Virgin and Child in a Walled Garden by the Master of the Triburtine Sibyl (£400k-£600k, above).

Image result for A Wide Village Street in summer with carts, villagers and gentlefolk sotheby's
A tiny Jan Brueghel the elder, A Wide Village Street in summer with carts, villagers and gentlefolk (the title says it all) reminds me that he's a really great artist (£2.5m-£3.5m, Sotheby's). Not all his pictures rise to this level, and weaker ones appear at auction quite often. It's a beautiful and easily appreciated picture, but it's also a sophisticated image. Perspective is cleverly distorted; compare the trees on the left and the right. It's a trick used by Rubens, but on a tiny scale. When you see lots of pictures of this type you come to appreciate how hard it is to integrate those seemingly-random figures into a harmonious whole. It's a really great picture.

Sotheby's has a sleeper in reverse. This Ecce Homo is described as Venetian School, early sixteenth century, with an estimate of £30k-£50k. The catalogue note doesn't mention that it was previously offered in New York in 2009 with full attribution to Lorenzo Lotto, endorsed by Keith Christiansen of the Met, with estimate of $400k-$600k. It's still a fine, unusual picture. I wonder if it would have been better marketed without the initial Lotto attribution, encouraging the trade to bid it up as a sleeper.
Image result for fuseli christie's arthur
There's a dearth of great drawings at this week's sales, but each auction house has a few spectacular things. There's an overwhelming early Fuseli at Christie's, The Faerie Queen appears to Prince Arthur (£150k-£250k, above). The most fascinating is a twenty-foot panorama of London just after the Napoleonic Wars, by Pierre Prévost (£200k-£300k at Sotheby's).

These are the big-money highlights, though not so big compared to contemporary art. I'll write a separate post tomorrow about the day sales and antiquities sales, where there are some really good pictures with really modest estimates.

Sunday, 3 December 2017

Christmas shopping: old masters at Sotheby's and Christie's

I hear there were some expensive things at recent contemporary sales, but I don't pay much attention to such vulgarities. The real action is at the Old Master auctions. They were queuing around the block last month at Christie's to see something from the contemporary sale, but few people visit the free, public views for the old master auctions.

I've always wanted to see a version of Titian's St Margaret that's in a lot of the Titian catalogues, listed as in the Kisters Collection, Kreuzlingen. Kisters pictures have been coming onto the market over the past few years, and the one I'd wanted to see is on preview at Sotheby's this week, ahead of its sale in New York in January. Alas, I was disappointed. There are flashes of brilliance, but it's got nothing on the Prado version. Rightly attributed to 'Titian and studio', the estimate of $2-3m tells you how much Titian is in it. Very rich people can bid anything to the sky if the both want it, and this might just attract the attention of gazillionaires. But $3m or so is a prudent estimate and a fair price for a big and striking picture with a bit of Titian in it.

Another Kisters picture is much more exciting: a Velazquez I didn't know at all! It's a collaboration with Pietro Martire Neri, but the face looks all Velazquez. Estimate is $3-4m, which seems cheap for just the face. You get the rest gratis. There's also a gorgeous double portrait attributed to Bartholomäus Zeitblom (surely too good for school, and who else?), and a Virgin and Child with St Anne by Holbein the Elder coming up in New York. Wowsers!

Winter 2017 is the Northern Renaissance season, largely thanks to the Kreuzlingen collection. Other than Cranach, these guys don't come to market often enough to estimate reliably. They're not especially fashionable, and for reasons I can't fathom museums don't generally buy them. In the past few decades the Getty has bought a few, the National Gallery in Washington has bought a wonderful Calvary by the wonderfully-named 'Master of the Death of St Nicholas of Münster' and UK museums have bought ... nichts. I shudder to think of what they could have had for the cost of all those British portraits.

This Betrayal of Christ is by an unknown master from the Lower Rhine, c. 1510-15. It's an exceptional and beautiful picture that is worthy of the National Gallery, estimated at £200k-£300k at Sotheby's. Lack of comparisons makes estimating it difficult, and good Northern pictures have tended to sell strongly recently. But for 'relative value' in the old master world, this is my pick. Tremendous quality, with such characterful figures.

Another potential bargain is a panel that might be by Van der Vliet at Christie's, which needs attention. I don't know the artist well enough to judge, but the quality looks high. A much larger version sold for a lot of money a few years ago. Rohan McCulloch pointed out on Twitter that this one made £5k at a regional auction not long ago, against a low estimate. It's interesting that the buyer flipped it to a major auction house without restoring it. That might mean that the buyer couldn't get an expert endorsement (seducing experts is a crucial skill in the art market). I think it's more likely that the buyer thought it the best way to maximise risk-adjusted return. Dealers are keen to buy 'sleepers', because it makes them look smart and the re-sale price won't be anchored by the auction price. The £12k-18k estimate is way too low if it's right. But cleaning it is a cost and a risk; it might not be as good as it seems, and it might not be accepted. Even if it is, there's a risk that the right buyers aren't there on the day, and trade buyers won't pay full price for a fully-authenticated work. If I were a cynic, I'd be tempted to under-attribute and tip off potential buyers as a deliberate marketing strategy.

Speaking of cynical marketing strategy, I wonder if this picture would sell better as three or four fragments. The catalogue entry says it's a clever exercise in distorted perspective, but I don't think the composition works. In real life it looks much better than the photo. I'm entranced by the details, like a camp proto-mannerist take on Andrea Mantegna. It's by the Master of the Figdor St Eustache, £300k-£400k at Sotheby's.

My favourite Christie's picture is early German, too. This Massacre of the Innocents is by the Master of the Dinkensbühl Altarpiece, another wonderful moniker (beaten only by the Master of the Kitten). Wonderfully theatrical characterisation of calculated thuggery overseen by a sociopathically impassive king, wicked dog, and distraught mothers. Estimate is just £400-£600k, which is surely within the reach of many museums. Christie's also has an exciting rediscovery: a rare picture by the Prague mannerist Bartholomäus Spranger, Mercury carrying Psyche to Mount Olympus (£400k-£600k). Tragically it's rather abraded, like so many of his pictures that were presumably looted during the Thirty Years' War. It's still a rare and important picture that would be a fine acquisition by a major museum like the Met or the National Gallery in Washington, representing a school of art that bloomed brightly but briefly.

As ever, there are eye-watering gems in the sub-£100k range, which barely buys a snotty handkerchief by the trendy contemporaries. Adriaen van Stalbemt will never be a household name, but this picture of The building of the Tabernacle with the Israelites sewing the curtains was good enough to be included as an Elsheimer in an old catalogue raisonné. It had been downgraded by 1977, when it sold for about £65k in inflation-adjusted terms. In 1977 inflation was nearly 16%, income tax was 83% above £20k and globalisation hadn't begun. Today it's estimated at £60k-£80k.

I'm not a great admirer of Victorian pictures, but this Landseer sketch is gorgeous, at £50k-£70k. I wish I knew more about sculpture. There seems to be a gap in the mid-market, with wealthy collectors paying millions for the very best, but bargains in the low-five-figure range. Sotheby's is previewing a tremendous collection of drawings including a Fra Bartolommeo landscape from the album sold in the 1950s and a couple of Watteaus. They're being sold in New York in January, and I'll say more about them then. In the meantime, I'm off to buy some lottery tickets.

Tuesday, 28 November 2017

How not to clean pictures: the risk of gels

An art dealer has taken down the video nasty of this picture being scrubbed 'clean' with a harsh brush while gel drips across the cleaned area. You can still see it here as of today. The American Institute for Conservation issued a critical statement and conservators were furious. I don't know who was cleaning the picture, but a conservator on social media calls him 'Scrubby Jelly Pants'. The vigorous scraping with a harsh brush is horrifying, and obviously risks damage. We'll never know how much harm was done; working that quickly makes assessment impossible. In the image above the uncleaned eyebrow looks clearer than the cleaned one, despite the varnish. Of course it's possible that it's an effect of light or they were touched up by a later restorer, but the most likely explanation is damage to the original paint. But although the risk of mechanical abrasion is obvious, my concern here is with the relatively new use of gels in conservation.

Gels allow better control of solvents, restricting penetration. Conservators worry that residues could continue to act on painting long after the restoration, but extensive research suggests that the risk is low. My concern is slightly different: it's that new techniques encourage over-confidence. The defence of Scrubby Jelly Pants has been to assert that gels can be tailored to remove only varnish. Bullish conservators have always claimed only to remove varnish; no one wants to say they're removing original paint. But it's just not true. There's always a risk of removing original paint, particularly when it's applied as glazes that might be made of pigment suspended in varnish. Gels are not magic, and cleaning pictures is not an exact science.

Psychologists speculate that we have a 'risk budget', so we respond to new safety measures by taking more risk. Make us wear seatbelts, and we'll drive faster. The other risk is that new techniques are over-sold. There's a mountain of research on the benefits and uses of gels in conservation, but none of it says you can mix a Magic Gel that can be safely slavered over a picture and scrubbed away without risk. One of the main benefits of gels is the controlled application of solvent; in this video it is so thin it runs down the surface. If it were truly the case that the solvent can only dissolve varnish, and not affect underlying paint, it would not be necessary to apply it in a controlled way. The whole point of gels is to mitigate that risk.

The Hippocratic Oath - "first, do no harm" - is not appropriate in conservation. It would mean doing nothing. Old pictures, unlike people, do not heal or regenerate. Over time we can only ever have less of the original artwork. The challenge is to minimise the damage and weigh it against the benefits. Conservation is essential and desirable; a great masterpiece hidden behind dirt and varnish can't be enjoyed by anyone. But the idea of doing irreversible damage to irreplaceable art is horrifying, so the temptation is to deny the trade-off. I recently heard the leader of a major conservation project assert that there was 'no risk' involved. That either misunderstands risk, or mistrusts her audience to understand the trade-off.

Conservation has become highly professionalised and sophisticated. Good conservators use artistic knowledge and skill in the application of advanced scientific techniques that are extensively tested and debated. But their work is not widely appreciated outside the field. There ought to be a better quality debate about the costs and benefits of conservation projects, rather than spin about 'zero risk, all benefit'. The failure to have that debate is part of what creates the space for videos like this to go viral, and for bad conservation techniques to thrive.

Monday, 21 August 2017

Museums are not the answer for Confederate monuments

Confederate monuments celebrate a failed war to defend slavery. Most were erected long after the Civil War as a deliberate assertion of white supremacy, alongside Jim Crow laws and a resurgent Ku Klux Klan. Other historical figures have flaws against the standards of our times, but statues of Washington and Jefferson celebrate their enlightened nation-building, not their slave ownership. I understand why so many think the statues must go. I'm still not persuaded. People really are worried about a slippery slope that will carry away all historic statues. They're not racists, and they're not fools. Radicals really are targeting Columbus now.

But this debate isn't really about the merits of Confederate monuments, and I don't think the statue smashers' primary concern is racism. In a recent poll a plurality of African Americans want the statues to remain. That's especially awkward for protesters who believe we should defer to minority experience, so they've mostly ignored it. The inflamed passions are really a manifestation of the culture wars. The most angry voices and most sanctimonious arguments are from the people who most strongly identify with one or other tribe in the culture wars. They're animated by hatred for the other side in the here and now, not racism. It's telling that the debate over symbols has been so much more gripping and more inspiring than policy debates about issues like gerrymandering or civil forfeiture or policing that have much more real-world impact here and now.

Any solution that doesn't recognise that will fail. The task isn't deciding what to do with statues, it's working out how to pacify culture warriors. And that's why I think some of the 'middle ground' solutions are the worst of all worlds.

Putting the statues in museums assumes that museums will give the 'right' interpretation, so that the oiks who won't 'get it' in the public square can be made to understand. It co-opts museums for a particular side in the culture war. Sorting institutions into 'ours' and 'theirs' is a catastrophic strategy. Museums already tend to lean liberal. Many were founded as patriotic projects, but today they're more likely to indulge the cringing political correctness of Fred Wilson. Putting the sculptures in museums won't calm the passions. It will shift the focus to museums, further politicising them and alienating a large section of the public.

Adding new statues of heroes that today's critics approve is also a problem. They are bound to be a focus of bitter debate when the symbolism is raised to such importance. Maybe they could raise statues of Frederick Douglass holding a rifle, to show his support for gun rights? And it feels condescending to require a statue of a black hero directly opposite a Confederate criminal to provide 'balance'. A non-trivial problem is that a lot of public sculpture is quite awful, and bad monuments have proliferated in the UK and sullied our towns and cities.

I don't have any better ideas. I think it's tragic that iconoclasm has become a radical strategy, and also tragic that so many conservatives struggle to concede the awfulness of the Confederacy. But anything that reinforces destructive tribal loyalties in the culture wars will make things worse.

Tuesday, 25 July 2017

It's all about context: assessing the old master market

The old master auctions that I wrote about recently did ... OK. Sold percentage was high, but the major Turner that I didn't care for just squeaked by at £18.5m. Some wonderful Northern pictures did deservedly well, a portrait of Anne of Hungary's court fool (above) that they gave to Jan Sanders van Hemessen making £2.2m against an upper estimate of £600k. A marvellous Murillo made £2.7m, a little less than it made in 2005, adjusted for inflation. Conventional wisdom is that the market doesn't like 'stale' pictures, but twelve years is enough of a gap for a new generation of collectors to come through. And whilst novelty doubtless has some inherent value, there are other reasons for recent returns to do poorly at auction. The person willing to bid highest last time has dropped out - because they're selling it. The last auction established an anchor price, so it's easier to offer around the market than something with uncertain value. Auction might be the last resort. And they might be selling because it wasn't as good as they hoped, after cleaning and research. Finally the market is inherently volatile. There just aren't that many people chasing after each lot, and indeed some pictures returning to auction do very well indeed, from this to this.

Art market reports tend to read to much into each auction. It's a small sample, and it's mostly noise rather than signal. The July sales were solid, but not spectacular, so people tended to read into them what they wanted. Some old master dealers are too keen on talking up their market. Short term fluctuations are market volatility are literally their living, but sometimes being too close to the action means missing the context. Just because some dealers might be making a killing doesn't mean the market is in splendid good health. So I find myself again in disagreement with Art History News.

Supply of old masters fluctuates a bit, but not by as much as you might think. Great things do still come to market, and there’s a fairly steady stream of material. But they’re not making any more, so it is a finite market. The key change is demand. Art and antiques are bought by the affluent and the rich. And their ranks have been multiplying. There’s been a massive growth of global wealth, and a particularly striking growth in the super-rich. The potential market has been growing.

If a population increases and grows richer, a car manufacturer with static sales shouldn’t get too excited. There are more potential customers, but they’re not buying cars. If it turns out that all the other manufacturers are selling more and more cars, as you’d expect in a growing market, our manufacturer ought to get a bit worried. That's the situation in the old masters market; other sectors in the art market are booming.

Bendor Grosvenor references a lightweight report by Arts Economics. It’s hard to assess the report because there are so few references. The lack of caveats (uncertainty about size of market given different definitions and different sales channels) makes it look more like a marketing brochure than serious research. But it’s still hard to read as an endorsement of the old master market. Old masters are ‘best performing’ in the UK market only in the context of relative increase (of 16%). But that’s against a decline of 50% the year before ($438m to $219m). Of course that’s partly driven by decisions about where to sell, but it shows the danger of cherry-picking data.

The report confuses regions and hubs, and is padded out with unsubstantiated claims about the “knowledge-intensive and gender-balanced” jobs that are provided (what is a ‘gender-balanced’ job?). But it does show that the old master market is almost the smallest segment of the fine art market: 45% post-war and contemporary, 30% modern, 12% impressionist and post-impressionist and just 13% old master. So the market post-war and contemporary – just a few generations – is nearly as large as the seven previous centuries.

Billionaires' net worth has increased roughly fivefold since 1995. Globalisation has created a vast new upper middle class in developing countries who are able to afford works of art. The boom in contemporary art isn't surprising. The remarkable thing is that so few rich people are spending their wealth on old masters.

It’s not a slight on old masters, or on the people who market them for a living, to say that the market is weak. I see it more as an indictment of the taste of the rich, but we shouldn’t take rich people’s taste too seriously. If you have even a little spare money you can buy pictures that really ought to be out of your league. Enjoy it while it lasts.